From 80fa24e1015cba902ca66d0a09e0a0ccd1fc87f5 Mon Sep 17 00:00:00 2001 From: Krishnaprasad Hande Date: Fri, 6 Sep 2024 10:31:35 -0700 Subject: [PATCH 1/4] Updating the VM SKU page with link to onboarding user guide --- .../sizes/gpu-accelerated/nccadsh100v5-series.md | 3 +++ 1 file changed, 3 insertions(+) diff --git a/articles/virtual-machines/sizes/gpu-accelerated/nccadsh100v5-series.md b/articles/virtual-machines/sizes/gpu-accelerated/nccadsh100v5-series.md index cc4d5a3a74b..e27dd947834 100644 --- a/articles/virtual-machines/sizes/gpu-accelerated/nccadsh100v5-series.md +++ b/articles/virtual-machines/sizes/gpu-accelerated/nccadsh100v5-series.md @@ -101,4 +101,7 @@ Accelerator (GPUs, FPGAs, etc.) info for each size --- +## Creating VM +To get started with NCCads H100 v5 VMs and for information regarding confidential GPU driver and attesetation configuration, see [confidential GPU Configuration](https://aka.ms/cgpu-onboarding-steps) + [!INCLUDE [sizes-footer](../includes/sizes-footer.md)] From b2f9590d8d1359a7cf08d2d70fd02121aa5ae6af Mon Sep 17 00:00:00 2001 From: Matt McInnes Date: Fri, 6 Sep 2024 14:25:50 -0700 Subject: [PATCH 2/4] Initial file removal and redirect modification --- ...blishing.redirection.virtual-machines.json | 130 +++++++++++ articles/virtual-machines/hb-series.md | 65 ------ articles/virtual-machines/hbv2-series.md | 65 ------ articles/virtual-machines/hbv3-series.md | 65 ------ articles/virtual-machines/hbv4-series.md | 63 ------ articles/virtual-machines/hc-series.md | 65 ------ articles/virtual-machines/hx-series.md | 65 ------ articles/virtual-machines/lasv3-series.md | 73 ------- articles/virtual-machines/lsv2-series.md | 91 -------- articles/virtual-machines/lsv3-series.md | 75 ------- .../virtual-machines/nc-a100-v4-series.md | 84 -------- articles/virtual-machines/nc-series.md | 70 ------ articles/virtual-machines/nct4-v3-series.md | 68 ------ articles/virtual-machines/ncv2-series.md | 78 ------- articles/virtual-machines/ncv3-series.md | 72 ------- .../virtual-machines/nd-h100-v5-series.md | 61 ------ articles/virtual-machines/nd-series.md | 69 ------ articles/virtual-machines/nda100-v4-series.md | 66 ------ .../virtual-machines/ndm-a100-v4-series.md | 74 ------- articles/virtual-machines/ndv2-series.md | 76 ------- .../virtual-machines/ngads-v-620-series.md | 68 ------ articles/virtual-machines/np-series.md | 201 ------------------ articles/virtual-machines/nv-series.md | 70 ------ articles/virtual-machines/nva10v5-series.md | 76 ------- articles/virtual-machines/nvv3-series.md | 66 ------ articles/virtual-machines/nvv4-series.md | 67 ------ .../sizes-b-series-burstable.md | 69 ------ 27 files changed, 130 insertions(+), 1962 deletions(-) delete mode 100644 articles/virtual-machines/hb-series.md delete mode 100644 articles/virtual-machines/hbv2-series.md delete mode 100644 articles/virtual-machines/hbv3-series.md delete mode 100644 articles/virtual-machines/hbv4-series.md delete mode 100644 articles/virtual-machines/hc-series.md delete mode 100644 articles/virtual-machines/hx-series.md delete mode 100644 articles/virtual-machines/lasv3-series.md delete mode 100644 articles/virtual-machines/lsv2-series.md delete mode 100644 articles/virtual-machines/lsv3-series.md delete mode 100644 articles/virtual-machines/nc-a100-v4-series.md delete mode 100644 articles/virtual-machines/nc-series.md delete mode 100644 articles/virtual-machines/nct4-v3-series.md delete mode 100644 articles/virtual-machines/ncv2-series.md delete mode 100644 articles/virtual-machines/ncv3-series.md delete mode 100644 articles/virtual-machines/nd-h100-v5-series.md delete mode 100644 articles/virtual-machines/nd-series.md delete mode 100644 articles/virtual-machines/nda100-v4-series.md delete mode 100644 articles/virtual-machines/ndm-a100-v4-series.md delete mode 100644 articles/virtual-machines/ndv2-series.md delete mode 100644 articles/virtual-machines/ngads-v-620-series.md delete mode 100644 articles/virtual-machines/np-series.md delete mode 100644 articles/virtual-machines/nv-series.md delete mode 100644 articles/virtual-machines/nva10v5-series.md delete mode 100644 articles/virtual-machines/nvv3-series.md delete mode 100644 articles/virtual-machines/nvv4-series.md delete mode 100644 articles/virtual-machines/sizes-b-series-burstable.md diff --git a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json index 1341bcb74c0..01d35f5f1e5 100644 --- a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json +++ b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json @@ -6364,6 +6364,136 @@ "source_path_from_root": "/articles/virtual-machines/esv6-edsv6-series.md", "redirect_url": "/azure/virtual-machines/sizes/memory-optimized/esv6-series", "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/sizes-b-series-burstable.md", + "redirect_url": "/azure/virtual-machines/sizes/general-purpose/bv1-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/lsv2-series.md", + "redirect_url": "/azure/virtual-machines/sizes/storage-optimized/lsv2-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/lsv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/storage-optimized/lsv3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/lasv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/storage-optimized/lasv3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nvv4-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nvv4-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nvv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nvv3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nva10v5-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nva10v5-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nv-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nv-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/np-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/np-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/ngads-v-620-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ngadsv620-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/ndv2-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ndv2-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/ndm-a100-v4-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ndma100v4-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nda100-v4-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nda100v4-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nd-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nd-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nd-h100-v5-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ndh100v5-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nvv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nvv3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/ncv2-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncv2-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nct4-v3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nct4v3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nc-a100-v4-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nca100v4-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/nc-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nc-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hx-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hx-series.", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hc-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hc-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hbv4-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hbv4-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hbv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hbv3-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hbv2-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hbv2-series", + "redirect_document_id": true + }, + { + "source_path_from_root": "/articles/virtual-machines/hb-series.md", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hb-series", + "redirect_document_id": true } ] } \ No newline at end of file diff --git a/articles/virtual-machines/hb-series.md b/articles/virtual-machines/hb-series.md deleted file mode 100644 index 4432611a26f..00000000000 --- a/articles/virtual-machines/hb-series.md +++ /dev/null @@ -1,65 +0,0 @@ ---- -title: HB-series -description: Specifications for the HB-series VMs. -ms.service: azure-virtual-machines -ms.subservice: hpc -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: cynthn -ms.author: padmalathas ---- - -# HB-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HB-series VMs are optimized for applications that are driven by memory bandwidth, such as fluid dynamics, explicit finite element analysis, and weather modeling. HB VMs feature 60 AMD EPYC 7551 processor cores, 4 GB of RAM per CPU core, and no simultaneous multithreading. An HB VM provides up to 260 GB/sec of memory bandwidth. - -HB-series VMs feature 100 Gb/sec Mellanox EDR InfiniBand. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. These VMs support Adaptive Routing and the Dynamic Connected Transport (DCT, in addition to standard RC and UD transports). These features enhance application performance, scalability, and consistency, and their usage is recommended. - -[ACU](acu.md): 199-216
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/accelerated-networking-on-hb-hc-hbv2-and-ndv2/ba-p/2067965) about performance and potential issues)
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-
- -| Size | vCPU | Processor | Memory (GiB) | Memory bandwidth GB/s | Base CPU frequency (GHz) | All-cores frequency (GHz, peak) | Single-core frequency (GHz, peak) | RDMA performance (Gb/s) | MPI support | Temp storage (GiB) | Max data disks | Max Ethernet vNICs | -| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_HB60rs | 60 | AMD EPYC 7551 | 228 | 263 | 2.0 | 2.55 | 2.55 | 100 | All | 700 | 4 | 8 | -| Standard_HB60-45rs | 45 | AMD EPYC 7551 | 228 | 263 | 2.0 | 2.55 | 2.55 | 100 | All | 700 | 4 | 8 | -| Standard_HB60-30rs | 30 | AMD EPYC 7551 | 228 | 263 | 2.0 | 2.55 | 2.55 | 100 | All | 700 | 4 | 8 | -| Standard_HB60-15rs | 15 | AMD EPYC 7551 | 228 | 263 | 2.0 | 2.55 | 2.55 | 100 | All | 700 | 4 | 8 | - -Learn more about the: -- [Architecture and VM topology](hb-series-overview.md) -- Supported [software stack](hb-series-overview.md#software-specifications) including supported OS -- Expected [performance](hb-series-performance.md) of the HB-series VM - -[!INCLUDE [hpc-include.md](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a higher level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/hbv2-series.md b/articles/virtual-machines/hbv2-series.md deleted file mode 100644 index 536c06c5c14..00000000000 --- a/articles/virtual-machines/hbv2-series.md +++ /dev/null @@ -1,65 +0,0 @@ ---- -title: HBv2-series - Azure Virtual Machines -description: Specifications for the HBv2-series VMs. -ms.service: azure-virtual-machines -ms.subservice: hpc -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: cynthn -ms.author: padmalathas ---- - -# HBv2-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HBv2-series VMs are optimized for applications that are driven by memory bandwidth, such as fluid dynamics, finite element analysis, and reservoir simulation. HBv2 VMs feature 120 AMD EPYC 7V12 processor cores, 4 GB of RAM per CPU core, and no simultaneous multithreading. Each HBv2 VM provides up to 350 GB/s of memory bandwidth, and up to 4 teraFLOPS of FP64 compute. - -HBv2-series VMs feature 200 Gb/sec Mellanox HDR InfiniBand. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. These VMs support Adaptive Routing and the Dynamic Connected Transport (DCT, in addition to standard RC and UD transports). These features enhance application performance, scalability, and consistency, and their usage is recommended. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/accelerated-networking-on-hb-hc-hbv2-and-ndv2/ba-p/2067965) about performance and potential issues)
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-
- -| Size | vCPU | Processor | Memory (GiB) | Memory bandwidth GB/s | Base CPU frequency (GHz) | All-cores frequency (GHz, peak) | Single-core frequency (GHz, peak) | RDMA performance (Gb/s) | MPI support | Temp storage (GiB) | Max data disks | Max Ethernet vNICs | -| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_HB120rs_v2 | 120 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 | -| Standard_HB120-96rs_v2 | 96 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 | -| Standard_HB120-64rs_v2 | 64 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 | -| Standard_HB120-32rs_v2 | 32 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 | -| Standard_HB120-16rs_v2 | 16 | AMD EPYC 7V12 | 456 | 350 | 2.45 | 3.1 | 3.3 | 200 | All | 480 + 960 | 8 | 8 | - -Learn more about the: -- [Architecture and VM topology](hbv2-series-overview.md) -- Supported [software stack](hbv2-series-overview.md#software-specifications) including supported OS -- Expected [performance](hbv2-performance.md) of the HBv2-series VM - -[!INCLUDE [hpc-include](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a higher level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/hbv3-series.md b/articles/virtual-machines/hbv3-series.md deleted file mode 100644 index 9ab16c01314..00000000000 --- a/articles/virtual-machines/hbv3-series.md +++ /dev/null @@ -1,65 +0,0 @@ ---- -title: HBv3-series - Azure Virtual Machines -description: Specifications for the HBv3-series VMs. -ms.service: azure-virtual-machines -ms.subservice: hpc -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: cynthn -ms.author: padmalathas ---- - -# HBv3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HBv3-series VMs are optimized for HPC applications such as fluid dynamics, explicit and implicit finite element analysis, weather modeling, seismic processing, reservoir simulation, and RTL simulation. HBv3 VMs feature up to 120 AMD EPYC™ 7V73X (Milan-X) CPU cores, 448 GB of RAM, and no simultaneous multithreading. HBv3-series VMs also provide 350 GB/sec of memory bandwidth (amplified up to 630 GB/s), up to 96 MB of L3 cache per core (1.536 GB total per VM), up to 7 GB/s of block device SSD performance, and clock frequencies up to 3.5 GHz. - -All HBv3-series VMs feature 200 Gb/sec HDR InfiniBand from NVIDIA Networking to enable supercomputer-scale MPI workloads. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. The HDR InfiniBand fabric also supports Adaptive Routing and the Dynamic Connected Transport (DCT, in additional to standard RC and UD transports). These features enhance application performance, scalability, and consistency, and their usage is strongly recommended. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Not supported
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/accelerated-networking-on-hb-hc-hbv2-and-ndv2/ba-p/2067965) about performance and potential issues)
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-
- -|Size |vCPU |Processor |Memory (GiB) |Memory bandwidth GB/s |Base CPU frequency (GHz) |All-cores frequency (GHz, peak) |Single-core frequency (GHz, peak) |RDMA performance (Gb/s) |MPI support |Temp storage (GiB) |Max data disks |Max Ethernet vNICs | -|----|----|----|----|----|----|----|----|----|----|----|----|----| -|Standard_HB120rs_v3 |120 |AMD EPYC 7V73X |448 |350 |1.9 |3.0 |3.5 |200 |All |2 * 960 |32 |8 | -|Standard_HB120-96rs_v3 |96 |AMD EPYC 7V73X |448 |350 |1.9 |3.0 |3.5 |200 |All |2 * 960 |32 |8 | -|Standard_HB120-64rs_v3 |64 |AMD EPYC 7V73X |448 |350 |1.9 |3.0 |3.5 |200 |All |2 * 960 |32 |8 | -|Standard_HB120-32rs_v3 |32 |AMD EPYC 7V73X |448 |350 |1.9 |3.0 |3.5 |200 |All |2 * 960 |32 |8 | -|Standard_HB120-16rs_v3 |16 |AMD EPYC 7V73X |448 |350 |1.9 |3.0 |3.5 |200 |All |2 * 960 |32 |8 | - -Learn more about the: -- [Architecture and VM topology](hbv3-series-overview.md) -- Supported [software stack](hbv3-series-overview.md#software-specifications) including supported OS -- Expected [performance](hbv3-performance.md) of the HBv3-series VM - -[!INCLUDE [hpc-include](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a higher level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/hbv4-series.md b/articles/virtual-machines/hbv4-series.md deleted file mode 100644 index cfe2b4d8720..00000000000 --- a/articles/virtual-machines/hbv4-series.md +++ /dev/null @@ -1,63 +0,0 @@ ---- -title: HBv4-series - Azure Virtual Machines -description: Specifications for the HBv4-series VMs. -ms.service: azure-virtual-machines -author: Padmalathas -ms.author: padmalathas -ms.subservice: hpc -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: wwilliams ---- - -# HBv4-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HBv4-series VMs are optimized for various HPC workloads such as computational fluid dynamics, finite element analysis, frontend and backend EDA, rendering, molecular dynamics, computational geoscience, weather simulation, and financial risk analysis. HBv4 VMs feature up to 176 AMD EPYC™ 9V33X ("Genoa-X") CPU cores with AMD's 3D V-Cache, clock frequencies up to 3.7 GHz, and no simultaneous multithreading. HBv4-series VMs also provide 768 GB of RAM, 2.3 GB L3 cache. The 2.3 GB L3 cache per VM can deliver up to 5.7 TB/s of bandwidth to amplify up to 780 GB/s of bandwidth from DRAM, for a blended average of 1.2 TB/s of effective memory bandwidth across a broad range of customer workloads. The VMs also provide up to 12 GB/s (reads) and 7 GB/s (writes) of block device SSD performance. - - -All HBv4-series VMs feature 400 Gb/s NDR InfiniBand from NVIDIA Networking to enable supercomputer-scale MPI workloads. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. NDR continues to support features like Adaptive Routing and the Dynamically Connected Transport (DCT). This newest generation of InfiniBand also brings greater support for offload of MPI collectives, optimized real-world latencies due to congestion control intelligence, and enhanced adaptive routing capabilities. These features enhance application performance, scalability, and consistency, and their usage is recommended. - -[Premium Storage](premium-storage-performance.md): Supported\ -[Premium Storage caching](premium-storage-performance.md): Supported\ -[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)\ -[Live Migration](maintenance-and-updates.md): Not Supported\ -[Memory Preserving Updates](maintenance-and-updates.md): Not Supported\ -[VM Generation Support](generation-2.md): Generation 2\ -[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli)\ -[Ephemeral OS Disks](ephemeral-os-disks.md): Supported -
- -|Size |Physical CPU cores |Processor |Memory (GB) |Memory bandwidth (GB/s) |Base CPU frequency (GHz) |Single-core frequency (GHz, peak) |RDMA performance (Gb/s) |MPI support |Temp storage (TB) |Max data disks |Max Ethernet vNICs | -|----|----|----|----|----|----|----|----|----|----|----|----| -|Standard_HB176rs_v4 |176 |AMD EPYC 9V33X (Genoa-X) |768 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HB176-144rs_v4|144 |AMD EPYC 9V33X (Genoa-X) |768 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HB176-96rs_v4 |96 |AMD EPYC 9V33X (Genoa-X) |768 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HB176-48rs_v4 |48 |AMD EPYC 9V33X (Genoa-X) |768 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HB176-24rs_v4 |24 |AMD EPYC 9V33X (Genoa-X) |768 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | - -[!INCLUDE [hpc-include](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples, and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a higher level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/hc-series.md b/articles/virtual-machines/hc-series.md deleted file mode 100644 index fdf93330552..00000000000 --- a/articles/virtual-machines/hc-series.md +++ /dev/null @@ -1,65 +0,0 @@ ---- -title: HC-series - Azure Virtual Machines -description: Specifications for the HC-series VMs. -ms.service: azure-virtual-machines -ms.subservice: hpc -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: wwilliams -ms.author: padmalathas ---- - -# HC-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HC-series VMs are optimized for applications driven by dense computation, such as implicit finite element analysis, molecular dynamics, and computational chemistry. HC VMs feature 44 Intel Xeon Platinum 8168 processor cores, 8 GB of RAM per CPU core, and no hyperthreading. The Intel Xeon Platinum platform supports Intel’s rich ecosystem of software tools such as the Intel Math Kernel Library and advanced vector processing capabilities such as AVX-512. - -HC-series VMs feature 100 Gb/sec Mellanox EDR InfiniBand. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. These VMs support Adaptive Routing and the Dynamic Connected Transport (DCT, in addition to standard RC and UD transports). These features enhance application performance, scalability, and consistency, and their usage is recommended. - -[ACU](acu.md): 297-315
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/accelerated-networking-on-hb-hc-hbv2-and-ndv2/ba-p/2067965) about performance and potential issues)
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-
- -| Size | vCPU | Processor | Memory (GiB) | Memory bandwidth GB/s | Base CPU frequency (GHz) | All-cores frequency (GHz, peak) | Single-core frequency (GHz, peak) | RDMA performance (Gb/s) | MPI support | Temp storage (GiB) | Max data disks | Max Ethernet vNICs | -| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_HC44rs | 44 | Intel Xeon Platinum 8168 | 352 | 191 | 2.7 | 3.4 | 3.7 | 100 | All | 700 | 4 | 8 | -| Standard_HC44-16rs | 16 | Intel Xeon Platinum 8168 | 352 | 191 | 2.7 | 3.4 | 3.7 | 100 | All | 700 | 4 | 8 | -| Standard_HC44-32rs | 32 | Intel Xeon Platinum 8168 | 352 | 191 | 2.7 | 3.4 | 3.7 | 100 | All | 700 | 4 | 8 | - - -Learn more about the: -- [Architecture and VM topology](hc-series-overview.md) -- Supported [software stack](hc-series-overview.md#software-specifications) including supported OS -- Expected [performance](hc-series-performance.md) of the HC-series VM - -[!INCLUDE [hpc-include](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a high-level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/hx-series.md b/articles/virtual-machines/hx-series.md deleted file mode 100644 index 2b149599e49..00000000000 --- a/articles/virtual-machines/hx-series.md +++ /dev/null @@ -1,65 +0,0 @@ ---- -title: HX-series - Azure Virtual Machines -description: Specifications for the HX-series VMs. -ms.service: azure-virtual-machines -ms.subservice: hpc -author: Padmalathas -ms.author: padmalathas -ms.topic: conceptual -ms.date: 07/25/2024 -ms.reviewer: wwilliams ---- - -# HX-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -HX-series VMs are optimized for workloads that require significant memory capacity with twice the memory capacity as HBv4. For example, workloads such as silicon design can use HX-series VMs to enable EDA customers targeting the most advanced manufacturing processes to run their most memory-intensive workloads. - -HX VMs feature up to 176 AMD EPYC™ 9V33X ("Genoa-X") CPU cores with AMD's 3D V-Cache, clock frequencies up to 3.7 GHz, and no simultaneous multithreading. HX-series VMs also provide 1.4 TB of RAM, 2.3 GB L3 cache. The 2.3 GB L3 cache per VM can deliver up to 5.7 TB/s of bandwidth to amplify up to 780 GB/s of bandwidth from DRAM, for a blended average of 1.2 TB/s of effective memory bandwidth across a broad range of customer workloads. The VMs also provide up to 12 GB/s (reads) and 7 GB/s (writes) of block device SSD performance. - -All HX-series VMs feature 400 Gb/s NDR InfiniBand from NVIDIA Networking to enable supercomputer-scale MPI workloads. These VMs are connected in a non-blocking fat tree for optimized and consistent RDMA performance. NDR continues to support features like Adaptive Routing and the Dynamically Connected Transport (DCT). This newest generation of InfiniBand also brings greater support for offload of MPI collectives, optimized real-world latencies due to congestion control intelligence, and enhanced adaptive routing capabilities. These features enhance application performance, scalability, and consistency, and their usage is recommended. - -[Premium Storage](premium-storage-performance.md): Supported\ -[Premium Storage caching](premium-storage-performance.md): Supported\ -[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)\ -[Live Migration](maintenance-and-updates.md): Not Supported\ -[Memory Preserving Updates](maintenance-and-updates.md): Not Supported\ -[VM Generation Support](generation-2.md): Generation 2\ -[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli)\ -[Ephemeral OS Disks](ephemeral-os-disks.md): Supported -
- -|Size |Physical CPU cores |Processor |Memory (GB) |Memory per core (GB) |Memory bandwidth (GB/s) |Base CPU frequency (GHz) |Single-core frequency (GHz, peak) |RDMA performance (Gb/s) |MPI support |Temp storage (TB) |Max data disks |Max Ethernet vNICs | -|----|----|----|----|----|----|----|----|----|----|----|----|----| -|Standard_HX176rs |176 |AMD EPYC 9V33X (Genoa-X) |1408 |8 |780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HX176-144rs|144 |AMD EPYC 9V33X (Genoa-X) |1408 |10|780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HX176-96rs |96 |AMD EPYC 9V33X (Genoa-X) |1408 |15|780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HX176-48rs |48 |AMD EPYC 9V33X (Genoa-X) |1408 |29|780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | -|Standard_HX176-24rs |24 |AMD EPYC 9V33X (Genoa-X) |1408 |59|780 |2.4 |3.7 |400 |All |2 * 1.8 |32 |8 | - - -[!INCLUDE [hpc-include](./includes/hpc-include.md)] - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - - -## Next steps - -- Read about the latest announcements, HPC workload examples and performance results at the [Azure Compute Tech Community Blogs](https://techcommunity.microsoft.com/t5/azure-compute/bg-p/AzureCompute). -- For a high-level architectural view of running HPC workloads, see [High Performance Computing (HPC) on Azure](/azure/architecture/topics/high-performance-computing/). -- Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/lasv3-series.md b/articles/virtual-machines/lasv3-series.md deleted file mode 100644 index b5a1bc20971..00000000000 --- a/articles/virtual-machines/lasv3-series.md +++ /dev/null @@ -1,73 +0,0 @@ ---- -title: Lasv3-series - Azure Virtual Machines -description: Specifications for the Lasv3-series of Azure Virtual Machines (Azure VMs). -author: sasha-melamed -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 06/01/2022 -ms.author: sasham ---- - -# Lasv3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The Lasv3-series of Azure Virtual Machines (Azure VMs) features high-throughput, low latency, directly mapped local NVMe storage. These VMs run on an AMD 3rd Generation EPYC™ 7763v processor in a multi-threaded configuration with an L3 cache of up to 256 MB that can achieve a boosted maximum frequency of 3.5 GHz. The Lasv3-series VMs are available in sizes from 8 to 80 vCPUs in a simultaneous multi-threading configuration. There are 8 GiB of memory per vCPU, and one 1.92 TB NVMe SSD device per 8 vCPUs, with up to 19.2 TB (10x1.92TB) available on the L80as_v3 size. - -> [!NOTE] -> The Lasv3-series VMs are optimized to use the local disk on the node attached directly to the VM rather than using [durable data disks](disks-types.md). This method allows for greater IOPS and throughput for your workloads. The Lsv3, Lasv3, Lsv2, and Ls-series don't support the creation of a local cache to increase the IOPS achievable by durable data disks. -> -> The high throughput and IOPS of the local disk makes the Lasv3-series VMs ideal for NoSQL stores such as Apache Cassandra and MongoDB. These stores replicate data across multiple VMs to achieve persistence in the event of the failure of a single VM. -> -> To learn more, see how optimize performance on Lasv3-series [Windows-based VMs](../virtual-machines/windows/storage-performance.md) or [Linux-based VMs](../virtual-machines/linux/storage-performance.md). - -- [Premium Storage](premium-storage-performance.md): Supported -- [Premium Storage caching](premium-storage-performance.md): Not Supported -- [Live Migration](maintenance-and-updates.md): Not Supported -- [Memory Preserving Updates](maintenance-and-updates.md): Supported -- [VM Generation Support](generation-2.md): Generation 1 and 2 -- [Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported -- [Ephemeral OS Disks](ephemeral-os-disks.md): Supported -- [Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Supported - -| Size | vCPU | Memory (GiB) | Temp disk (GiB) | NVMe Disks | NVMe Disk throughput (Read IOPS/MBps) | Uncached data disk throughput (IOPS/MBps) | Max burst uncached data disk throughput (IOPS/MBps)| Max Data Disks | Max NICs | Expected network bandwidth (Mbps) | -|---|---|---|---|---|---|---|---|---|---|---| -| Standard_L8as_v3 | 8 | 64 | 80 | 1x1.92 TB | 400000/2000 | 12800/200 | 20000/1280 | 16 | 4 | 12500 | -| Standard_L16as_v3 | 16 | 128 | 160 | 2x1.92 TB | 800000/4000 | 25600/384 | 40000/1280 | 32 | 8 | 12500 | -| Standard_L32as_v3 | 32 | 256 | 320 | 4x1.92 TB | 1.5M/8000 | 51200/768 | 80000/1600 | 32 | 8 | 16000 | -| Standard_L48as_v3 | 48 | 384 | 480 | 6x1.92 TB | 2.2M/14000 | 76800/1152 | 80000/2000 | 32 | 8 | 24000 | -| Standard_L64as_v3 | 64 | 512 | 640 | 8x1.92 TB | 2.9M/16000 | 80000/1280 | 80000/2000 | 32 | 8 | 32000 | -| Standard_L80as_v3 | 80 | 640 | 800 | 10x1.92TB | 3.8M/20000 | 80000/1400 | 80000/2000 | 32 | 8 | 32000 | - -1. **Temp disk**: Lasv3-series VMs have a standard SCSI-based temp resource disk for use by the OS paging or swap file (`D:` on Windows, `/dev/sdb` on Linux). This disk provides 80 GiB of storage, 4000 IOPS, and 80 MBps transfer rate for every 8 vCPUs. For example, Standard_L80as_v3 provides 800 GiB at 40000 IOPS and 800 MBPS. This configuration ensures that the NVMe drives can be fully dedicated to application use. This disk is ephemeral, and all data is lost on stop or deallocation. -2. **NVMe Disks**: NVMe disk throughput can go higher than the specified numbers. However, higher performance isn't guaranteed. Local NVMe disks are ephemeral. Data is lost on these disks if you stop or deallocate your VM. -3. **NVMe Disk encryption** Lasv3 VMs created or allocated on or after 1/1/2023 have their local NVMe drives encrypted by default using hardware-based encryption with a Platform-managed key, except for the regions listed below. - -> [!NOTE] -> Central US and Qatar Central Lasv3 VMs created or allocated on or after 4/1/2023 have their local NVMe Drives Encrypted. - -4. **NVMe Disk throughput**: Hyper-V NVMe Direct technology provides unthrottled access to local NVMe drives mapped securely into the guest VM space. Lasv3 NVMe disk throughput can go higher than the specified numbers, but higher performance isn't guaranteed. To achieve maximum performance, see how to optimize performance on Lasv3-series [Windows-based VMs](../virtual-machines/windows/storage-performance.md) or [Linux-based VMs](../virtual-machines/linux/storage-performance.md). Read/write performance varies based on IO size, drive load, and capacity utilization. -5. **Max burst uncached data disk throughput**: Lasv3-series VMs can [burst their disk performance](./disk-bursting.md) for up to 30 minutes at a time. - -> [!NOTE] -> Lasv3-series VMs don't provide a host cache for the data disk because this configuration doesn't benefit the Lasv3 workloads. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -More information on Disks Types: [Disk Types](./disks-types.md#ultra-disks) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/lsv2-series.md b/articles/virtual-machines/lsv2-series.md deleted file mode 100644 index a16e440d568..00000000000 --- a/articles/virtual-machines/lsv2-series.md +++ /dev/null @@ -1,91 +0,0 @@ ---- -title: Lsv2-series - Azure Virtual Machines -description: Specifications for the Lsv2-series VMs. -author: sasha-melamed -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.custom: -ms.topic: conceptual -ms.date: 06/01/2022 -ms.author: sasham ---- - -# Lsv2-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Uniform scale sets - -The Lsv2-series features high throughput, low latency, directly mapped local NVMe storage running on the [AMD EPYCTM 7551 processor](https://www.amd.com/en/processors/epyc-server-cpu-family) with an all core boost of 2.55GHz and a max boost of 3.0GHz. The Lsv2-series VMs come in sizes from 8 to 80 vCPU in a simultaneous multi-threading configuration. There is 8 GiB of memory per vCPU, and one 1.92TB NVMe SSD M.2 device per 8 vCPUs, with up to 19.2TB (10x1.92TB) available on the L80s v2. - -> [!NOTE] -> The Lsv2-series VMs are optimized to use the local disk on the node attached directly to the VM rather than using durable data disks. This allows for greater IOPs / throughput for your workloads. The Lsv2 and Ls-series do not support the creation of a local cache to increase the IOPs achievable by durable data disks. -> -> The high throughput and IOPs of the local disk makes the Lsv2-series VMs ideal for NoSQL stores such as Apache Cassandra and MongoDB which replicate data across multiple VMs to achieve persistence in the event of the failure of a single VM. -> -> To learn more, see Optimize performance on the Lsv2-series virtual machines for [Windows](../virtual-machines/windows/storage-performance.md) or [Linux](../virtual-machines/linux/storage-performance.md). - -[ACU](acu.md): 150-175
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Not Supported
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-Bursting: Supported
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory (GiB) | Temp disk1 (GiB) | NVMe Disks2 | NVMe Disk throughput3 (Read IOPS/MBps) | Uncached data disk throughput (IOPs/MBps)4 | Max burst uncached data disk throughput (IOPs/MBps)5| Max Data Disks | Max NICs | Expected network bandwidth (Mbps) | -|---|---|---|---|---|---|---|---|---|---|---| -| Standard_L8s_v2 | 8 | 64 | 80 | 1x1.92 TB | 400000/2000 | 8000/160 | 8000/1280 | 16 | 2 | 3200 | -| Standard_L16s_v2 | 16 | 128 | 160 | 2x1.92 TB | 800000/4000 | 16000/320 | 16000/1280 | 32 | 4 | 6400 | -| Standard_L32s_v2 | 32 | 256 | 320 | 4x1.92 TB | 1.5M/8000 | 32000/640 | 32000/1280 | 32 | 8 | 12800 | -| Standard_L48s_v2 | 48 | 384 | 480 | 6x1.92 TB | 2.2M/14000 | 48000/960 | 48000/2000 | 32 | 8 | 16000+ | -| Standard_L64s_v2 | 64 | 512 | 640 | 8x1.92 TB | 2.9M/16000 | 64000/1280 | 64000/2000 | 32 | 8 | 16000+ | -| Standard_L80s_v26 | 80 | 640 | 800 | 10x1.92TB | 3.8M/20000 | 80000/1400 | 80000/2000 | 32 | 8 | 16000+ | - -1 Lsv2-series VMs have a standard SCSI based temp resource disk for OS paging/swap file use (D: on Windows, /dev/sdb on Linux). This disk provides 80 GiB of storage, 4,000 IOPS, and 80 MBps transfer rate for every 8 vCPUs (e.g. Standard_L80s_v2 provides 800 GiB at 40,000 IOPS and 800 MBPS). This ensures the NVMe drives can be fully dedicated to application use. This disk is Ephemeral, and all data will be lost on stop/deallocate. - -2 Local NVMe disks are ephemeral, data will be lost on these disks if you stop/deallocate your VM. Local NVMe disks aren't encrypted by [Azure Storage encryption](disk-encryption.md), even if you enable [encryption at host](disk-encryption.md#supported-vm-sizes). - -3 Hyper-V NVMe Direct technology provides unthrottled access to local NVMe drives mapped securely into the guest VM space. Achieving maximum performance requires using either the latest WS2019 build or Ubuntu 18.04 or 16.04 from the Azure Marketplace. Write performance varies based on IO size, drive load, and capacity utilization. - -4 Lsv2-series VMs do not provide host cache for data disk as it does not benefit the Lsv2 workloads. - -5 Lsv2-series VMs can [burst](./disk-bursting.md) their disk performance for up to 30 minutes at a time. - -6 VMs with more than 64 vCPUs require one of these supported guest operating systems: - -- Windows Server 2016 or later -- Ubuntu 18.04 LTS or later -- SLES 12 SP5 or later -- RHEL 6.10, with Microsoft-provided LIS package 4.3.1 (or later) installed -- RHEL 7.9 or later -- Oracle Linux with UEK4 or later -- Debian 9 with the backports kernel, Debian 10 or later - -## Size table definitions - -- Storage capacity is shown in units of GiB or 1024^3 bytes. When comparing disks measured in GB (1000^3 bytes) to disks measured in GiB (1024^3) remember that capacity numbers given in GiB may appear smaller. For example, 1023 GiB = 1098.4 GB -- Disk throughput is measured in input/output operations per second (IOPS) and MBps where MBps = 10^6 bytes/sec. -- If you want to get the best performance for your VMs, you should limit the number of data disks to 2 disks per vCPU. -- **Expected network bandwidth** is the maximum aggregated [bandwidth allocated per VM type](/azure/virtual-network/virtual-machine-network-throughput) across all NICs, for all destinations. Upper limits are not guaranteed, but are intended to provide guidance for selecting the right VM type for the intended application. Actual network performance will depend on a variety of factors including network congestion, application loads, and network settings. For information on optimizing network throughput, see [Optimizing network throughput for Windows and Linux](/azure/virtual-network/virtual-network-optimize-network-bandwidth). To achieve the expected network performance on Linux or Windows, it may be necessary to select a specific version or optimize your VM. For more information, see [How to reliably test for virtual machine throughput](/azure/virtual-network/virtual-network-bandwidth-testing). - - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -More information on Disks Types: [Disk Types](./disks-types.md#ultra-disks) - - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/lsv3-series.md b/articles/virtual-machines/lsv3-series.md deleted file mode 100644 index 5538699e90e..00000000000 --- a/articles/virtual-machines/lsv3-series.md +++ /dev/null @@ -1,75 +0,0 @@ ---- -title: Lsv3-series - Azure Virtual Machines -description: Specifications for the Lsv3-series of Azure Virtual Machines (Azure VMs). -author: sasha-melamed -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 06/01/2022 -ms.author: sasham ---- - -# Lsv3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The Lsv3-series of Azure Virtual Machines (Azure VMs) features high-throughput, low latency, directly mapped local NVMe storage. These VMs run on the 3rd Generation Intel® Xeon® Platinum 8370C (Ice Lake) processor in a [hyper-threaded configuration](https://www.intel.com/content/www/us/en/architecture-and-technology/hyper-threading/hyper-threading-technology.html). This new processor features an all-core turbo clock speed of 3.5 GHz with [Intel® Turbo Boost Technology](https://www.intel.com/content/www/us/en/architecture-and-technology/turbo-boost/turbo-boost-technology.html), [Intel® Advanced-Vector Extensions 512 (Intel® AVX-512)](https://www.intel.com/content/www/us/en/architecture-and-technology/avx-512-overview.html) and [Intel® Deep Learning Boost](https://software.intel.com/content/www/us/en/develop/topics/ai/deep-learning-boost.html). - -The Lsv3-series VMs are available in sizes from 8 to 80 vCPUs. There are 8 GiB of memory allocated per vCPU, and one 1.92TB NVMe SSD device allocated per 8 vCPUs, with up to 19.2TB (10x1.92TB) available on the L80s_v3 size. - -> [!NOTE] -> The Lsv3-series VMs are optimized to use the local disk on the node attached directly to the VM rather than using [durable data disks](disks-types.md). This method allows for greater IOPS and throughput for your workloads. The Lsv3, Lasv3, Lsv2, and Ls-series VMs don't support the creation of a host cache to increase the IOPS achievable by durable data disks. -> -> The high throughput and IOPS of the local disk makes the Lsv3-series VMs ideal for NoSQL stores such as Apache Cassandra and MongoDB. These stores replicate data across multiple VMs to achieve persistence in the event of the failure of a single VM. -> -> To learn more, see how to optimize performance on the Lsv3-series [Windows-based VMs](../virtual-machines/windows/storage-performance.md) or [Linux-based VMs](../virtual-machines/linux/storage-performance.md). - -- [Premium Storage](premium-storage-performance.md): Supported -- [Premium Storage caching](premium-storage-performance.md): Not Supported -- [Live Migration](maintenance-and-updates.md): Not Supported -- [Memory Preserving Updates](maintenance-and-updates.md): Supported -- [VM Generation Support](generation-2.md): Generation 1 and 2 -- [Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported -- [Ephemeral OS Disks](ephemeral-os-disks.md): Supported -- [Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Supported - -| Size | vCPU | Memory (GiB) | Temp disk (GiB) | NVMe Disks | NVMe Disk throughput (Read IOPS/MBps) | Uncached data disk throughput (IOPS/MBps) | Max burst uncached data disk throughput (IOPS/MBps) | Max Data Disks | Max NICs | Expected network bandwidth (Mbps) | -|---|---|---|---|---|---|---|---|---|---|---| -| Standard_L8s_v3 | 8 | 64 | 80 | 1x1.92 TB | 400000/2000 | 12800/290 | 20000/1200 | 16 | 4 | 12500 | -| Standard_L16s_v3 | 16 | 128 | 160 | 2x1.92 TB | 800000/4000 | 25600/600 | 40000/1600 | 32 | 8 | 12500 | -| Standard_L32s_v3 | 32 | 256 | 320 | 4x1.92 TB | 1.5M/8000 | 51200/865 | 80000/2000 | 32 | 8 | 16000 | -| Standard_L48s_v3 | 48 | 384 | 480 | 6x1.92 TB | 2.2M/14000 | 76800/1315 | 80000/3000 | 32 | 8 | 24000 | -| Standard_L64s_v3 | 64 | 512 | 640 | 8x1.92 TB | 2.9M/16000 | 80000/1735 | 80000/3000 | 32 | 8 | 30000 | -| Standard_L80s_v3 | 80 | 640 | 800 | 10x1.92TB | 3.8M/20000 | 80000/2160 | 80000/3000 | 32 | 8 | 32000 | - -1. **Temp disk**: Lsv3-series VMs have a standard SCSI-based temp resource disk for use by the OS paging or swap file (`D:` on Windows, `/dev/sdb` on Linux). This disk provides 80 GiB of storage, 4,000 IOPS, and 80 MBps transfer rate for every 8 vCPUs. For example, Standard_L80s_v3 provides 800 GiB at 40000 IOPS and 800 MBPS. This configuration ensures the NVMe drives can be fully dedicated to application use. This disk is ephemeral, and all data is lost on stop or deallocation. -2. **NVMe Disks**: NVMe disk throughput can go higher than the specified numbers. However, higher performance isn't guaranteed. Local NVMe disks are ephemeral. Data is lost on these disks if you stop or deallocate your VM. -3. **NVMe Disk encryption** Lsv3 VMs created or allocated on or after 1/1/2023 have their local NVMe drives encrypted by default using hardware-based encryption with a Platform-managed key, except for the regions listed below. - -> [!NOTE] -> Central US and Qatar Central Lsv3 VMs created or allocated on or after 4/1/2023 have their local NVMe Drives Encrypted. - -4. **NVMe Disk throughput**: Hyper-V NVMe Direct technology provides unthrottled access to local NVMe drives mapped securely into the guest VM space. Lsv3 NVMe disk throughput can go higher than the specified numbers, but higher performance isn't guaranteed. To achieve maximum performance, see how to optimize performance on the Lsv3-series [Windows-based VMs](../virtual-machines/windows/storage-performance.md) or [Linux-based VMs](../virtual-machines/linux/storage-performance.md). Read/write performance varies based on IO size, drive load, and capacity utilization. -5. **Max burst uncached data disk throughput**: Lsv3-series VMs can [burst their disk performance](./disk-bursting.md) for up to 30 minutes at a time. - -> [!NOTE] -> Lsv3-series VMs don't provide host cache for data disk as it doesn't benefit the Lsv3 workloads. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -More information on Disks Types: [Disk Types](./disks-types.md#ultra-disks) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nc-a100-v4-series.md b/articles/virtual-machines/nc-a100-v4-series.md deleted file mode 100644 index 3f3bf61648d..00000000000 --- a/articles/virtual-machines/nc-a100-v4-series.md +++ /dev/null @@ -1,84 +0,0 @@ ---- -title: NC A100 v4-series -description: Specifications for the NC A100 v4-series Azure VMs. These VMs include Linux, Windows, Flexible scale sets, and uniform scale sets.``` -author: sherrywangms -ms.author: sherrywang -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.custom: linux-related-content -ms.topic: conceptual -ms.date: 09/19/2023 ---- - -# NC A100 v4-series - -> [!CAUTION] -> This article references CentOS, a Linux distribution that is End Of Life (EOL) status. Please consider your use and plan accordingly. For more information, see the [CentOS End Of Life guidance](~/articles/virtual-machines/workloads/centos/centos-end-of-life.md). - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NC A100 v4 series virtual machine (VM) is a new addition to the Azure GPU family. You can use this series for real-world Azure Applied AI training and batch inference workloads. - -The NC A100 v4 series is powered by NVIDIA A100 PCIe GPU and third generation AMD EPYC™ 7V13 (Milan) processors. The VMs feature up to 4 NVIDIA A100 PCIe GPUs with 80 GB memory each, up to 96 non-multithreaded AMD EPYC Milan processor cores and 880 GiB of system memory. -These VMs are ideal for real-world Applied AI workloads, such as: - -- GPU-accelerated analytics and databases -- Batch inferencing with heavy pre- and post-processing -- Autonomy model training -- Oil and gas reservoir simulation -- Machine learning (ML) development -- Video processing -- AI/ML web services - - - -## Supported features - -To get started with NC A100 v4 VMs, refer to [HPC Workload Configuration and Optimization](configure.md) for steps including driver and network configuration. - -Due to increased GPU memory I/O footprint, the NC A100 v4 requires the use of [Generation 2 VMs](generation-2.md) and marketplace images. While the [Azure HPC images](configure.md) are recommended, Azure HPC Ubuntu 20.04 and Azure HPC CentOS 7.9, RHEL 8.8, RHEL 9.2, Windows Server 2019, and Windows Server 2022 images are supported. - -- [Premium Storage](premium-storage-performance.md): Supported -- [Premium Storage caching](premium-storage-performance.md): Supported -- [Ultra Disks](disks-types.md#ultra-disks): Not Supported -- [Live Migration](maintenance-and-updates.md): Not Supported -- [Memory Preserving Updates](maintenance-and-updates.md): Not Supported -- [VM Generation Support](generation-2.md): Generation 2 -- [Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported -- [Ephemeral OS Disks](ephemeral-os-disks.md): Supported -- InfiniBand: Not Supported -- NVIDIA NVLink Interconnect: Supported -- [Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported - - - -| Size | vCPU | Memory (GiB) | Temp Disk1 (GiB) | NVMe Disks2 | GPU3 | GPU Memory (GiB) | Max data disks | Max uncached disk throughput (IOPS / MBps) | Max NICs/network bandwidth (MBps) | -|---|---|---|---|---|---|---|---|---|---| -| Standard_NC24ads_A100_v4 | 24 | 220 |64 | 960 GB | 1 | 80 | 8 | 30000/1000 | 2/20,000 | -| Standard_NC48ads_A100_v4 | 48 | 440 | 128| 2x960 GB| 2 | 160 | 16 | 60000/2000 | 4/40,000 | -| Standard_NC96ads_A100_v4 | 96 | 880 | 256| 4x960 GB | 4 | 320 | 32 | 120000/4000 | 8/80,000 | - -1 NC A100 v4 series VMs have a standard SCSI based temp resource disk for OS paging/swap file use. This ensures the NVMe drives can be fully dedicated to application use. This disk is Ephemeral, and all data will be lost on stop/deallocate. - -2 Local NVMe disks are ephemeral, data will be lost on these disks if you stop/deallocate your VM. Local NVMe disk is coming as RAM and it needs to be manually formatted in newly deployed VM. - -31 GPU = one A100 80GB PCIe GPU card
- -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -You can [use the pricing calculator](https://azure.microsoft.com/pricing/calculator/) to estimate your Azure VMs costs. - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next step - -- [Compare compute performance across Azure SKUs with Azure compute units (ACU)](acu.md) diff --git a/articles/virtual-machines/nc-series.md b/articles/virtual-machines/nc-series.md deleted file mode 100644 index cedde7254c4..00000000000 --- a/articles/virtual-machines/nc-series.md +++ /dev/null @@ -1,70 +0,0 @@ ---- -title: NC-series - Azure Virtual Machines -description: Specifications for the NC-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 12/21/2022 -ms.author: vikancha ---- - -# NC-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -> [!IMPORTANT] -> NC and NC_Promo series Azure virtual machines (VMs) will be retired on September 6, 2023. For more information, see the [NC and NC_Promo retirement information](nc-series-retirement.md). For how to migrate your workloads to other VM sizes, see the [GPU compute migration guide](n-series-migration.md). -> -> This retirement announcement doesn't apply to NCv3, NCasT4v3 and NC A100 v4 series VMs. - -NC-series VMs are powered by the [NVIDIA Tesla K80](https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/tesla-product-literature/Tesla-K80-BoardSpec-07317-001-v05.pdf) card and the Intel Xeon E5-2690 v3 (Haswell) processor. Users can crunch through data faster by using CUDA for energy exploration applications, crash simulations, ray traced rendering, deep learning, and more. The NC24r configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads. - -[Premium Storage](premium-storage-performance.md): Not Supported
-[Premium Storage caching](premium-storage-performance.md): Not Supported
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Not Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Not Supported
-Nvidia NVLink Interconnect: Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs | -|---|---|---|---|---|---|---|---| -| Standard_NC6 | 6 | 56 | 340 | 1 | 12 | 24 | 1 | -| Standard_NC12 | 12 | 112 | 680 | 2 | 24 | 48 | 2 | -| Standard_NC24 | 24 | 224 | 1440 | 4 | 48 | 64 | 4 | -| Standard_NC24r* | 24 | 224 | 1440 | 4 | 48 | 64 | 4 | - -1 GPU = one-half K80 card. - -*RDMA capable - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nct4-v3-series.md b/articles/virtual-machines/nct4-v3-series.md deleted file mode 100644 index e2685268674..00000000000 --- a/articles/virtual-machines/nct4-v3-series.md +++ /dev/null @@ -1,68 +0,0 @@ ---- -title: NCas T4 v3-series -description: Specifications for the NCas T4 v3-series VMs. -ms.service: azure-virtual-machines -ms.subservice: sizes -author: vikancha-MSFT -ms.topic: conceptual -ms.date: 01/12/2021 -ms.author: vikancha ---- - -# NCasT4_v3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NCasT4_v3-series virtual machines are powered by [Nvidia Tesla T4](https://www.nvidia.com/en-us/data-center/tesla-t4/) GPUs and AMD EPYC 7V12(Rome) CPUs. The VMs feature up to 4 NVIDIA T4 GPUs with 16 GB of memory each, up to 64 non-multithreaded AMD EPYC 7V12 (Rome) processor cores(base frequency of 2.45 GHz, all-cores peak frequency of 3.1 GHz and single-core peak frequency of 3.3 GHz) and 440 GiB of system memory. These virtual machines are ideal for deploying AI services- such as real-time inferencing of user-generated requests, or for interactive graphics and visualization workloads using NVIDIA's GRID driver and virtual GPU technology. Standard GPU compute workloads based around CUDA, TensorRT, Caffe, ONNX and other frameworks, or GPU-accelerated graphical applications based on OpenGL and DirectX can be deployed economically, with close proximity to users, on the NCasT4_v3 series. - -
- -[ACU](acu.md): 230-260
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage, and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-Nvidia NVLink Interconnect: Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs / Expected network bandwidth (Mbps) | -| --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_NC4as_T4_v3 |4 |28 |176 | 1 | 16 | 8 | 2 / 8000 | -| Standard_NC8as_T4_v3 |8 |56 |352 | 1 | 16 | 16 | 4 / 8000 | -| Standard_NC16as_T4_v3 |16 |110 |352 | 1 | 16 | 32 | 8 / 8000 | -| Standard_NC64as_T4_v3 |64 |440 |2816 | 4 | 64 | 32 | 8 / 32000 | - - - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure NCasT4_v3-series VMs running Windows or Linux, Nvidia GPU drivers must be installed. - -To install Nvidia GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -The Azure Nvidia GPU driver extension will deploy CUDA drivers on the NCasT4_v3-series VMs. For graphics and visualization workloads manually install the GRID drivers supported by Azure. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/ncv2-series.md b/articles/virtual-machines/ncv2-series.md deleted file mode 100644 index 3e8ab00478c..00000000000 --- a/articles/virtual-machines/ncv2-series.md +++ /dev/null @@ -1,78 +0,0 @@ ---- -title: NCv2-series - Azure Virtual Machines -description: Specifications for the NCv2-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 02/03/2020 -ms.author: vikancha ---- - -# NCv2-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - - -> [!IMPORTANT] -> NCv2 series Azure virtual machines (VMs) will be retired on September 6, 2023. For more information, see the [NCv2 retirement information](ncv2-series-retirement.md). For how to migrate your workloads to other VM sizes, see the [GPU compute migration guide](n-series-migration.md). -> -> This retirement announcement doesn't apply to NCv3, NCasT4v3 and NC A100 v4 series VMs. - -NCv2-series VMs are powered by NVIDIA Tesla P100 GPUs. These GPUs can provide more than 2x the computational performance of the NC-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. In addition to the GPUs, the NCv2-series VMs are also powered by Intel Xeon E5-2690 v4 (Broadwell) CPUs. - -The NC24rs v2 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Not Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-Nvidia NVLink Interconnect: Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
- -> [!IMPORTANT] -> For this VM series, the vCPU (core) quota in your subscription is initially set to 0 in each region. [Request a vCPU quota increase](/azure/azure-portal/supportability/regional-quota-requests) for this series in an [available region](https://azure.microsoft.com/regions/services/). -> -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs | -|---|---|---|---|---|---|---|---|---| -| Standard_NC6s_v2 | 6 | 112 | 736 | 1 | 16 | 12 | 20000/200 | 4 | -| Standard_NC12s_v2 | 12 | 224 | 1474 | 2 | 32 | 24 | 40000/400 | 8 | -| Standard_NC24s_v2 | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 | -| Standard_NC24rs_v2* | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 | - -One GPU = one P100 card. - -*RDMA capable - - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/ncv3-series.md b/articles/virtual-machines/ncv3-series.md deleted file mode 100644 index 72ff66de01b..00000000000 --- a/articles/virtual-machines/ncv3-series.md +++ /dev/null @@ -1,72 +0,0 @@ ---- -title: NCv3-series - Azure Virtual Machines -description: Specifications for the NCv3-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 12/20/2023 -ms.author: vikancha ---- - -# NCv3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -NCv3-series VMs are powered by NVIDIA Tesla V100 GPUs. These GPUs can provide 1.5x the computational performance of the NCv2-series. Customers can take advantage of these updated GPUs for traditional HPC workloads such as reservoir modeling, DNA sequencing, protein analysis, Monte Carlo simulations, and others. The NC24rs v3 configuration provides a low latency, high-throughput network interface optimized for tightly coupled parallel computing workloads. In addition to the GPUs, the NCv3-series VMs are also powered by Intel Xeon E5-2690 v4 (Broadwell) CPUs. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2*
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Not Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-Nvidia NVLink Interconnect: Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
- ->*Standard_NC24rs_v3 is not supported on Generation 2 VMs. - -> [!IMPORTANT] -> For this VM series, the vCPU (core) quota in your subscription is initially set to 0 in each region. [Request a vCPU quota increase](/azure/azure-portal/supportability/regional-quota-requests) for this series in an [available region](https://azure.microsoft.com/regions/services/). These SKUs aren't available to trial or Visual Studio Subscriber Azure subscriptions. Your subscription level might not support selecting or deploying these SKUs. -> - -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs | -|---|---|---|---|---|---|---|---|---| -| Standard_NC6s_v3 | 6 | 112 | 736 | 1 | 16 | 12 | 20000/200 | 4 | -| Standard_NC12s_v3 | 12 | 224 | 1474 | 2 | 32 | 24 | 40000/400 | 8 | -| Standard_NC24s_v3 | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 | -| Standard_NC24rs_v3* | 24 | 448 | 2948 | 4 | 64 | 32 | 80000/800 | 8 | - -1 GPU = one V100 card. - -*RDMA capable - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nd-h100-v5-series.md b/articles/virtual-machines/nd-h100-v5-series.md deleted file mode 100644 index a98c55d2dba..00000000000 --- a/articles/virtual-machines/nd-h100-v5-series.md +++ /dev/null @@ -1,61 +0,0 @@ ---- -title: ND H100 v5-series -description: Specifications for the ND H100 v5-series VMs -author: iamwilliew -ms.author: wwilliams -ms.service: azure-virtual-machines -ms.topic: conceptual -ms.date: 08/04/2023 ---- - -# ND H100 v5-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - - -The ND H100 v5 series virtual machine (VM) is a new flagship addition to the Azure GPU family. It’s designed for high-end Deep Learning training and tightly coupled scale-up and scale-out Generative AI and HPC workloads. - -The ND H100 v5 series starts with a single VM and eight NVIDIA H100 Tensor Core GPUs. ND H100 v5-based deployments can scale up to thousands of GPUs with 3.2Tb/s of interconnect bandwidth per VM. Each GPU within the VM is provided with its own dedicated, topology-agnostic 400 Gb/s NVIDIA Quantum-2 CX7 InfiniBand connection. These connections are automatically configured between VMs occupying the same virtual machine scale set, and support GPUDirect RDMA. - -Each GPU features NVLINK 4.0 connectivity for communication within the VM, and the instance is backed by 96 physical 4th Gen Intel Xeon Scalable processor cores. - -These instances provide excellent performance for many AI, ML, and analytics tools that support GPU acceleration ‘out-of-the-box,’ such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. Additionally, the scale-out InfiniBand interconnect is supported by a large set of existing AI and HPC tools that are built on NVIDIA’s NCCL communication libraries for seamless clustering of GPUs. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra disk](disks-types.md#ultra-disks): Supported [Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage, and performance
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-Infiniband: Supported, GPUDirect RDMA, 8x400 Gigabit NDR
-NVIDIA NVLink Interconnect: Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- ->[!IMPORTANT] ->To get started with ND H100 v5 VMs, refer to HPC Workload Configuration and Optimization for steps including driver and network configuration. Due to increased GPU memory I/O footprint, the ND H100 v5 requires the use of Generation 2 VMs and marketplace images. -> ->Azure supports Ubuntu 20.04/22.04, RHEL 7.9/8.7/9.3, AlmaLinux 8.8/9.2, and SLES 15 for ND H100 v5 VMs. On Azure marketplace, there are offerings of optimized and pre-configured [Linux VM images](configure.md#vm-images) for HPC/AI workloads with a variety of HPC tools and libraries installed, and thus they are strongly recommended. Currently, Ubuntu-HPC 20.04/22.04 and AlmaLinux-HPC 8.6/8.7 VM images are supported. - -## Example -[comment]: # (The ND H100 v5 series supports the following kernel version: Ubuntu 20.04: 5.4.0-1046-azure) - -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU Memory GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max network bandwidth | Max NICs | -|---------------------|------|------------|------------------------|----------------------------|----------------|----------------|-----------------------------------------|------------------------------|----------| -| Standard_ND96isr_H100_v5 | 96 | 1900 | 28000 | 8 H100 80 GB GPUs(NVLink) | 80 | 32 | 40800/612 | 80,000 Mbps | 8 | - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - - -More information on Disks Types: [Disk Types](./disks-types.md#ultra-disks) diff --git a/articles/virtual-machines/nd-series.md b/articles/virtual-machines/nd-series.md deleted file mode 100644 index e4694e17ebf..00000000000 --- a/articles/virtual-machines/nd-series.md +++ /dev/null @@ -1,69 +0,0 @@ ---- -title: ND-series - Azure Virtual Machines -description: Specifications for the ND-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 12/20/2022 -ms.author: vikancha ---- - -# ND-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The ND-series virtual machines are a new addition to the GPU family designed for AI, and Deep Learning workloads. They offer excellent performance for training and inference. ND instances are powered by [NVIDIA Tesla P40](https://images.nvidia.com/content/pdf/tesla/184427-Tesla-P40-Datasheet-NV-Final-Letter-Web.pdf) GPUs and Intel Xeon E5-2690 v4 (Broadwell) CPUs. These instances provide excellent performance for single-precision floating point operations, for AI workloads utilizing Microsoft Cognitive Toolkit, TensorFlow, Caffe, and other frameworks. The ND-series also offers a much larger GPU memory size (24 GB), enabling to fit much larger neural net models. Like the NC-series, the ND-series offers a configuration with a secondary low-latency, high-throughput network through RDMA, and InfiniBand connectivity so you can run large-scale training jobs spanning many GPUs. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Not Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-Nvidia NVLink Interconnect: Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
- -> [!IMPORTANT] -> For this VM series, the vCPU (core) quota per region in your subscription is initially set to 0. [Request a vCPU quota increase](/azure/azure-portal/supportability/regional-quota-requests) for this series in an [available region](https://azure.microsoft.com/regions/services/). -> -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs | -|---|---|---|---|---|---|---|---|---| -| Standard_ND6s | 6 | 112 | 736 | 1 | 24 | 12 | 20000/200 | 4 | -| Standard_ND12s | 12 | 224 | 1474 | 2 | 48 | 24 | 40000/400 | 8 | -| Standard_ND24s | 24 | 448 | 2948 | 4 | 96 | 32 | 80000/800 | 8 | -| Standard_ND24rs* | 24 | 448 | 2948 | 4 | 96 | 32 | 80000/800 | 8 | - -1 GPU = one P40 card. - -*RDMA capable - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nda100-v4-series.md b/articles/virtual-machines/nda100-v4-series.md deleted file mode 100644 index a10da579935..00000000000 --- a/articles/virtual-machines/nda100-v4-series.md +++ /dev/null @@ -1,66 +0,0 @@ ---- -title: ND A100 v4-series -description: Specifications for the ND A100 v4-series VMs. -ms.service: azure-virtual-machines -ms.subservice: sizes -author: sherrywangms -ms.author: sherrywang -ms.topic: conceptual -ms.date: 03/13/2023 ---- - -# ND A100 v4-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets. - -The ND A100 v4 series virtual machine(VM) is a new flagship addition to the Azure GPU family. It's designed for high-end Deep Learning training and tightly coupled scale-up and scale-out HPC workloads. - -The ND A100 v4 series starts with a single VM and eight NVIDIA Ampere A100 40GB Tensor Core GPUs. ND A100 v4-based deployments can scale up to thousands of GPUs with an 1.6 TB/s of interconnect bandwidth per VM. Each GPU within the VM is provided with its own dedicated, topology-agnostic 200 GB/s NVIDIA Mellanox HDR InfiniBand connection. These connections are automatically configured between VMs occupying the same VM scale set, and support GPUDirect RDMA. - -Each GPU features NVLINK 3.0 connectivity for communication within the VM, and the instance is backed by 96 physical 2nd-generation AMD Epyc™ 7V12 (Rome) CPU cores. - -These instances provide excellent performance for many AI, ML, and analytics tools that support GPU acceleration 'out-of-the-box,' such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. Additionally, the scale-out InfiniBand interconnect is supported by a large set of existing AI and HPC tools that are built on NVIDIA's NCCL2 communication libraries for seamless clustering of GPUs. - -> [!IMPORTANT] -> To get started with ND A100 v4 VMs, refer to [HPC Workload Configuration and Optimization](configure.md) for steps including driver and network configuration. -> Due to increased GPU memory I/O footprint, the ND A100 v4 requires the use of [Generation 2 VMs](generation-2.md) and marketplace images. -> -> Azure supports Ubuntu 20.04/22.04, RHEL 7.9/8.7/9.3, AlmaLinux 8.8/9.2, and SLES 15 for ND A100 v4 VMs. On Azure marketplace, there are offerings of optimized and pre-configured [Linux VM images](configure.md#vm-images) for HPC/AI workloads with a variety of HPC tools and libraries installed, and thus they are strongly recommended. Currently, Ubuntu-HPC 20.04/22.04 and AlmaLinux-HPC 8.6/8.7 VM images are supported. - -
- -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage, and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[InfiniBand](./extensions/enable-infiniband.md): Supported, GPUDirect RDMA, 8 x 200 Gigabit HDR
-NVIDIA NVLink Interconnect: Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp Storage (SSD): GiB | GPU | GPU Memory: GiB | Max data disks | Max uncached disk throughput: IOPS / MBps | Max network bandwidth | Max NICs | -|---|---|---|---|---|---|---|---|---|---| -| Standard_ND96asr_A100_v4 | 96 | 900 | 6000 | 8 A100 40-GB GPUs (NVLink 3.0) | 320 | 32 | 80,000 / 800 | 24,000 Mbps | 8 | - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/ndm-a100-v4-series.md b/articles/virtual-machines/ndm-a100-v4-series.md deleted file mode 100644 index 1e122c595a2..00000000000 --- a/articles/virtual-machines/ndm-a100-v4-series.md +++ /dev/null @@ -1,74 +0,0 @@ ---- -title: NDm A100 v4-series -description: Specifications for the NDm A100 v4-series VMs. -author: sherrywangms -ms.author: sherrywang -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 03/13/2023 - ---- - -# NDm A100 v4-series - -> [!CAUTION] -> This article references CentOS, a Linux distribution that is End Of Life (EOL) status. Please consider your use and plan accordingly. For more information, see the [CentOS End Of Life guidance](~/articles/virtual-machines/workloads/centos/centos-end-of-life.md). - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NDm A100 v4 series virtual machine(VM) is a new flagship addition to the Azure GPU family. It's designed for high-end Deep Learning training and tightly coupled scale-up and scale-out HPC workloads. - -The NDm A100 v4 series starts with a single VM and eight NVIDIA Ampere A100 80GB Tensor Core GPUs. NDm A100 v4-based deployments can scale up to thousands of GPUs with an 1.6 TB/s of interconnect bandwidth per VM. Each GPU within the VM is provided with its own dedicated, topology-agnostic 200 GB/s NVIDIA Mellanox HDR InfiniBand connection. These connections are automatically configured between VMs occupying the same VM scale set, and support GPUDirect RDMA. - -Each GPU features NVLINK 3.0 connectivity for communication within the VM, and the instance is backed by 96 physical 2nd-generation AMD Epyc™ 7V12 (Rome) CPU cores. - -These instances provide excellent performance for many AI, ML, and analytics tools that support GPU acceleration 'out-of-the-box,' such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. Additionally, the scale-out InfiniBand interconnect is supported by a large set of existing AI and HPC tools that are built on NVIDIA's NCCL2 communication libraries for seamless clustering of GPUs. - -> [!IMPORTANT] -> To get started with NDm A100 v4 VMs, refer to [HPC Workload Configuration and Optimization](configure.md) for steps including driver and network configuration. -> Due to increased GPU memory I/O footprint, the NDm A100 v4 requires the use of [Generation 2 VMs](generation-2.md) and marketplace images. The [Azure HPC images](configure.md) are strongly recommended. Azure HPC Ubuntu 18.04, 20.04 and Azure HPC CentOS 7.9 images are supported. -> - -
- -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage, and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-InfiniBand: Supported, GPUDirect RDMA, 8 x 200 Gigabit HDR
-Nvidia NVLink Interconnect: Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
-The NDm A100 v4 series supports the following kernel versions:
-CentOS 7.9 HPC: 3.10.0-1160.24.1.el7.x86_64
-Ubuntu 18.04: 5.4.0-1043-azure
-Ubuntu 20.04: 5.4.0-1046-azure
-
- -| Size | vCPU | Memory: GiB | Temp Storage (SSD): GiB | GPU | GPU Memory: GiB | Max data disks | Max uncached disk throughput: IOPS / MBps | Max network bandwidth | Max NICs | -|---|---|---|---|---|---|---|---|---|---| -| Standard_ND96amsr_A100_v4 | 96 | 1900 | 6400 | 8 A100 80 GB GPUs (NVLink 3.0) | 80 | 32 | 80,000 / 800 | 24,000 Mbps | 8 | - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/ndv2-series.md b/articles/virtual-machines/ndv2-series.md deleted file mode 100644 index c425fcd867a..00000000000 --- a/articles/virtual-machines/ndv2-series.md +++ /dev/null @@ -1,76 +0,0 @@ ---- -title: NDv2-series -description: Specifications for the NDv2-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 12/20/2022 -ms.author: vikancha ---- - -# Updated NDv2-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NDv2-series virtual machine is a new addition to the GPU family designed for the needs of the most demanding GPU-accelerated AI, machine learning, simulation, and HPC workloads. - -NDv2 is powered by 8 NVIDIA Tesla V100 NVLINK-connected GPUs, each with 32 GB of GPU memory. Each NDv2 VM also has 40 non-HyperThreaded Intel Xeon Platinum 8168 (Skylake) cores and 672 GiB of system memory. - -NDv2 instances provide excellent performance for HPC and AI workloads utilizing CUDA GPU-optimized computation kernels, and the many AI, ML, and analytics tools that support GPU acceleration 'out-of-box,' such as TensorFlow, Pytorch, Caffe, RAPIDS, and other frameworks. - -Critically, the NDv2 is built for both computationally intense scale-up (harnessing 8 GPUs per VM) and scale-out (harnessing multiple VMs working together) workloads. The NDv2 series now supports 100-Gigabit InfiniBand EDR backend networking, similar to that available on the HB series of HPC VM, to allow high-performance clustering for parallel scenarios including distributed training for AI and ML. This backend network supports all major InfiniBand protocols, including those employed by NVIDIA’s NCCL2 libraries, allowing for seamless clustering of GPUs. - -> [!IMPORTANT] -> When [enabling InfiniBand](./workloads/hpc/enable-infiniband.md) on the ND40rs_v2 VM, please use the 4.7-1.0.0.1 Mellanox OFED driver. -> -> Due to increased GPU memory, the new ND40rs_v2 VM requires the use of [Generation 2 VMs](./generation-2.md) and marketplace images. -> -> Please note: The ND40s_v2 featuring 16 GB of per-GPU memory is no longer available for preview and has been superceded by the updated ND40rs_v2. - -
- -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-InfiniBand: Supported
-Nvidia NVLink Interconnect: Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp Storage (SSD): GiB | GPU | GPU Memory: GiB | Max data disks | Max uncached disk throughput: IOPS / MBps | Max network bandwidth | Max NICs | -|---|---|---|---|---|---|---|---|---|---| -| Standard_ND40rs_v2 | 40 | 672 | 2948 | 8 V100 32 GB (NVLink) | 32 | 32 | 80000 / 800 | 24000 Mbps | 8 | - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-linux.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md). - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/ngads-v-620-series.md b/articles/virtual-machines/ngads-v-620-series.md deleted file mode 100644 index 144342c9330..00000000000 --- a/articles/virtual-machines/ngads-v-620-series.md +++ /dev/null @@ -1,68 +0,0 @@ ---- -title: Overview of NGads V620 Series (preview) -description: Overview of NGads V620 series GPU-enabled virtual machines -author: isgonzalez-MSFT -ms.author: isgonzalez -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 06/11/2023 - ---- - -# NGads V620-series - - -The NGads V620 series are GPU-enabled virtual machines with CPU, memory resources and storage resources balanced to generate and stream high quality graphics for a high performance, interactive gaming experience hosted in Azure. They're powered by [AMD Radeon(tm) PRO V620 GPU](https://www.amd.com/en/newsroom/press-releases/2021-11-4-amd-radeon-pro-v620-gpu-delivers-powerful-multi-.html) and [AMD EPYC 7763 (Milan) CPUs](https://www.amd.com/en/products/cpu/amd-epyc-7763). The AMD Radeon PRO V620 GPUs have a maximum frame buffer of 32 GB, which can be divided up to four ways through hardware partitioning. The AMD EPYC CPUs have a base clock speed of 2.45 GHz and a boost1 speed of 3.5Ghz. VMs are assigned full cores instead of threads, enabling full access to AMD’s powerful “Zen 3” cores.
-(1 EPYC-018: Max boost for AMD EPYC processors is the maximum frequency achievable by any single core on the processor under normal operating conditions for server systems.) - -NGads instances come in four sizes, allowing customers to right-size their gaming environments for the performance and cost that best fits their business needs. The NG-series virtual machines feature partial GPUs to enable you to pick the right-sized virtual machine for GPU accelerated graphics applications and virtual desktops. The vm sizes start with 1/4 of a GPU with 8-GiB frame buffer up to a full GPU with 32-GiB frame buffer. The NGads VMs also feature Direct Disk NVMe ranging from 1 to 4x 960 GB disks per VM. - - -[Premium Storage](premium-storage-performance.md): Supported
-Premium Storage v2: Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU1 | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU Memory GiB2 | Max data disks | Max uncached disk throughput: IOPS/MBps | Direct Disk NVMe3 | Max NICs / Max network bandwidth (Mbps) | -|---|---|---|---|---|---|---|---|---|---| -| Standard_NG8ads_V620_v1 | 8 | 16 | 256 | 1/4 | 8 | 8 | 12800 / 200 | 1x 960 GB | 2 / 10000 | -| Standard_NG16ads_V620_v1 | 16 | 32 | 512 | 1/2 | 16 | 16 | 25600 / 384 | 2x 960 GB | 4 / 20000 | -| Standard_NG32ads_V620_v1 | 32 | 64 | 1024 | 1 | 32 | 32 | 51200 / 768 | 4x 960 GB | 8 / 40000 | -| Standard_NG32adms_V620_v1 | 32 | 176 | 1024 | 1 | 32 | 32 | 51200 / 768 | 4x 960 GB | 8 / 40000 | - -1 Physical Cores. -2 The actual GPU VRAM reported in the operating system will be little less due to Error Correcting Code (ECC) support.
-3 Local NVMe disks are ephemeral. Data is lost on these disks if you stop/deallocate your VM. Local NVMe disks aren't encrypted by Azure Storage encryption, even if you enable encryption at host. - -## Supported operating systems and drivers - -The NGads V620 Series VMs support a new AMD Cloud Software driver that comes in two editions: A Gaming driver with regular updates to support the latest titles, and a Professional driver for accelerated Virtual Desktop environments, with Radeon PRO optimizations to support high-end workstation applications.
-To take advantage of the GPU capabilities of Azure NGads V620 Series VMs, AMD GPU drivers must be installed. NG virtual machines currently support only Windows guest operating systems.
-To install AMD GPU drivers manually, see [N-series AMD GPU driver setup for Windows](./windows/n-series-amd-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : Not available during preview. - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/np-series.md b/articles/virtual-machines/np-series.md deleted file mode 100644 index ee5ee6f8b61..00000000000 --- a/articles/virtual-machines/np-series.md +++ /dev/null @@ -1,201 +0,0 @@ ---- -title: NP-series - Azure Virtual Machines -description: Specifications for the NP-series VMs. -author: charest -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 03/07/2023 -ms.author: marccharest ---- - -# NP-series - -> [!CAUTION] -> This article references CentOS, a Linux distribution that is End Of Life (EOL) status. Please consider your use and plan accordingly. For more information, see the [CentOS End Of Life guidance](~/articles/virtual-machines/workloads/centos/centos-end-of-life.md). - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NP-series virtual machines are powered by [Xilinx U250](https://www.xilinx.com/products/boards-and-kits/alveo/u250.html) FPGAs for accelerating workloads including machine learning inference, video transcoding, and database search & analytics. NP-series VMs are also powered by Intel Xeon 8171M (Skylake) CPUs with all core turbo clock speed of 3.2 GHz. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-VM Generation Support: Generation 1
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | FPGA | FPGA memory: GiB | Max data disks | Max NICs/ Expected network bandwidth (Mbps) | -|---|---|---|---|---|---|---|---| -| Standard_NP10s | 10 | 168 | 736 | 1 | 64 | 8 | 1 / 7500 | -| Standard_NP20s | 20 | 336 | 1474 | 2 | 128 | 16 | 2 / 15000 | -| Standard_NP40s | 40 | 672 | 2948 | 4 | 256 | 32 | 4 / 30000 | - - - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - - -## Frequently asked questions - -**Q:** How to request quota for NP VMs? - -**A:** Follow this page [Increase VM-family vCPU quotas](/azure/azure-portal/supportability/per-vm-quota-requests). NP VMs are available in East US, West US2, SouthCentral US, West Europe, SouthEast Asia, Japan East, and Canada Central. - -**Q:** What version of Vitis should I use? - -**A:** Xilinx recommends [Vitis 2022.1](https://www.xilinx.com/products/design-tools/vitis/vitis-platform.html), you can also use the Development VM marketplace options (Vitis 2022.1 Development VM for Ubuntu 18.04, Ubuntu 20.04, and CentOS 7.8) - -**Q:** Do I need to use NP VMs to develop my solution? - -**A:** No, you can develop on-premises and deploy to the cloud. Make sure to follow the [attestation documentation](./field-programmable-gate-arrays-attestation.md) to deploy on NP VMs. - -**Q:** What shell version is supported and how can I get the development files? - -**A:** The FPGAs in Azure NP VMs support Xilinx Shell 2.1 (gen3x16-xdma-shell_2.1). See Xilinx Page [Xilinx/Azure with Alveo U250](https://www.amd.com/en/where-to-buy/accelerators/alveo/cloud-solutions/microsoft-azure.html) to get the development shell files. - -**Q:** Which file returned from attestation should I use when programming my FPGA in an NP VM? - -**A:** Attestation returns two xclbins, **design.bit.xclbin** and **design.azure.xclbin**. Use **design.azure.xclbin**. - -**Q:** Where should I get all the XRT / Platform files? - -**A:** Visit Xilinx's [Microsoft-Azure](https://www.amd.com/en/where-to-buy/accelerators/alveo/cloud-solutions/microsoft-azure.html) site for all files. - -**Q:** What Version of XRT should I use? - -**A:** xrt_202210.2.13.479 - -**Q:** What is the target deployment platform? - -**A:** Use the following platforms. -- xilinx-u250-gen3x16-xdma-platform-2.1-3_all -- xilinx-u250-gen3x16-xdma-validate_2.1-3005608.1 - -**Q:** Which platform should I target for development? - -**A:** xilinx-u250-gen3x16-xdma-2.1-202010-1-dev_1-2954688_all - -**Q:** What are the supported Operating Systems? - -**A:** Xilinx and Microsoft have validated Ubuntu 18.04 LTS, Ubuntu 20.04 LTS, and CentOS 7.8. - ->Xilinx has created the following marketplace images to simplify the deployment of these VMs: -> ->- Xilinx Alveo U250 2022.1 Deployment VM [Ubuntu18.04](https://portal.azure.com/#create/xilinx.vitis2022_1_ubuntu1804_development_imagevitis2022_1_ubuntu1804) -> ->- Xilinx Alveo U250 2022.1 Deployment VM [Ubuntu20.04](https://portal.azure.com/#create/xilinx.vitis2022_1_ubuntu2004_development_imagevitis2022_1_ubuntu2004) -> ->- Xilinx Alveo U250 2022.1 Deployment VM [CentOS7.8](https://portal.azure.com/#create/xilinx.vitis2022_1_centos78_development_imagevitis2022_1_centos78) - -**Q:** Can I deploy my Own Ubuntu / CentOS VMs and install XRT / Deployment Target Platform? - -**A:** Yes. - -**Q:** If I deploy my own Ubuntu18.04 VM then what are the required packages and steps? - -**A:** Follow the guidance in Xilinx XRT documentation [Xilinx XRT documentation](https://docs.xilinx.com/r/en-US/ug1451-xrt-release-notes/XRT-Operating-System-Support) - ->Install the following packages. ->- xrt_202210.2.13.479_18.04-amd64-xrt.deb -> ->- xrt_202210.2.13.479_18.04-amd64-azure.deb -> ->- xilinx-u250-gen3x16-xdma-platform-2.1-3_all_18.04.deb.tar.gz -> ->- xilinx-u250-gen3x16-xdma-validate_2.1-3005608.1_all.deb - -**Q:** If I deploy my own Ubuntu20.04 VM then what are the required packages and steps? - -**A:** Follow the guidance in Xilinx XRT documentation [Xilinx XRT documentation](https://docs.xilinx.com/r/en-US/ug1451-xrt-release-notes/XRT-Operating-System-Support) - ->Install the following packages. ->- xrt_202210.2.13.479_20.04-amd64-xrt.deb -> ->- xrt_202210.2.13.479_20.04-amd64-azure.deb -> ->- xilinx-u250-gen3x16-xdma-platform-2.1-3_all_18.04.deb.tar.gz -> ->- xilinx-u250-gen3x16-xdma-validate_2.1-3005608.1_all.deb - -**Q:** If I deploy my own CentOS7.8 VM then what are the required packages and steps? - -**A:** Follow the guidance in Xilinx XRT documentation [Xilinx XRT documentation](https://docs.xilinx.com/r/en-US/ug1451-xrt-release-notes/XRT-Operating-System-Support) - - >Install the following packages. - >- xrt_202210.2.13.479_7.8.2003-x86_64-xrt.rpm - > - >- xrt_202210.2.13.479_7.8.2003-x86_64-azure.rpm - > - >- xilinx-u250-gen3x16-xdma-platform-2.1-3.noarch.rpm.tar.gz - > - >- xilinx-u250-gen3x16-xdma-validate-2.1-3005608.1.noarch.rpm - -**Q:** What are the differences between on-premise FPGAs and NP VMs? - -**A:** -
-- Regarding XOCL/XCLMGMT: -
-On Azure NP VMs, only the role endpoint (Device ID 5005), which uses the XOCL driver, is present. - -In on-premise FPGAs, both the management endpoint (Device ID 5004) and role endpoint (Device ID 5005), which use the XCLMGMT and XOCL drivers respectively, are present. - -
-- Regarding XRT: -
-On Azure NP VMs, the XDMA 2.1 platform only supports Host_Mem(SB). -
-To enable Host_Mem(SB) (up to 1-Gb RAM): sudo xbutil host_mem --enable --size 1g -
-To disable Host_Mem(SB): sudo xbutil host_mem --disable -
- -
-Starting on XRT2021.1: - -On-premise FPGA in Linux exposes -[M2M data transfer](https://xilinx.github.io/XRT/master/html/m2m.html). -
-This feature isn't supported in Azure NP VMs. - -**Q:** Can I run xbmgmt commands? - -**A:** No, on Azure VMs there's no management support directly from the Azure VM. - - **Q:** Do I need to load a PLP? - -**A:** No, the PLP is loaded automatically for you, so there's no need to load via xbmgmt commands. - - -**Q:** Does Azure support different PLPs? - -**A:** Not at this time. We only support the PLP provided in the deployment platform packages. - -**Q:** How can I query the PLP information? - -**A:** Need to run xbutil query and look at the lower portion. - -**Q:** Do Azure NP VMs support FPGA bitstreams with Networking GT Kernel connections? - -**A:** No. The FPGA Attestation service performs a series of validations on a design checkpoint file and will generate an error if the user's application contains connections to the FPGA card's QSFP networking ports. - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nv-series.md b/articles/virtual-machines/nv-series.md deleted file mode 100644 index 0aa18d64d97..00000000000 --- a/articles/virtual-machines/nv-series.md +++ /dev/null @@ -1,70 +0,0 @@ ---- -title: NV-series - Azure Virtual Machines -description: Specifications for the NV-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 03/29/2022 -ms.author: vikancha ---- - -# NV-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - - -> [!IMPORTANT] -> NV and NV_Promo series Azure virtual machines (VMs) will be retired on September 6, 2023. For more information, see the [NV and NV_Promo retirement information](nv-series-retirement.md). For how to migrate your workloads to other VM sizes, see the [NV and NV_Promo series migration guide](nv-series-migration-guide.md). -> -> This retirement announcement doesn't apply to NVv3 and NVv4 series VMs. - -The NV-series virtual machines are powered by [NVIDIA Tesla M60](https://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology for desktop accelerated applications and virtual desktops where customers are able to visualize their data or simulations. Users are able to visualize their graphics intensive workflows on the NV instances to get superior graphics capability and additionally run single precision workloads such as encoding and rendering. NV-series VMs are also powered by Intel Xeon E5-2690 v3 (Haswell) CPUs. - -Each GPU in NV instances comes with a GRID license. This license gives you the flexibility to use an NV instance as a virtual workstation for a single user, or 25 concurrent users can connect to the VM for a virtual application scenario. - -[Premium Storage](premium-storage-performance.md): Not Supported
-[Premium Storage caching](premium-storage-performance.md): Not Supported
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Not Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Not Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max NICs | Virtual Workstations | Virtual Applications | -|---|---|---|---|---|---|---|---|---|---| -| Standard_NV6 | 6 | 56 | 340 | 1 | 8 | 24 | 1 | 1 | 25 | -| Standard_NV12 | 12 | 112 | 680 | 2 | 16 | 48 | 2 | 2 | 50 | -| Standard_NV24 | 24 | 224 | 1440 | 4 | 32 | 64 | 4 | 4 | 100 | - -1 GPU = one-half M60 card. - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nva10v5-series.md b/articles/virtual-machines/nva10v5-series.md deleted file mode 100644 index 749e448e790..00000000000 --- a/articles/virtual-machines/nva10v5-series.md +++ /dev/null @@ -1,76 +0,0 @@ ---- -title: NV A10 v5-series -description: Specifications for the NV A10 v5-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 02/01/2022 -ms.author: vikancha -ms.custom: references_regions ---- - -# NVadsA10 v5-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NVadsA10v5-series virtual machines are powered by [NVIDIA A10](https://www.nvidia.com/en-us/data-center/products/a10-gpu/) GPUs and AMD EPYC 74F3V(Milan) CPUs with a base frequency of 3.2 GHz, all-cores peak frequency of 4.0 GHz. With NVadsA10v5-series Azure is introducing virtual machines with partial NVIDIA GPUs. Pick the right sized virtual machine for GPU accelerated graphics applications and virtual desktops starting at 1/6th of a GPU with 4-GiB frame buffer to a full A10 GPU with 24-GiB frame buffer. - -Each virtual machine instance in NVadsA10v5-series comes with a GRID license. This license gives you the flexibility to use an NV instance as a virtual workstation for a single user, or 25 concurrent users can connect to the VM for a virtual application scenario. - -
- -[ACU](acu.md): Not Available
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU partition | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs / Expected network bandwidth (Mbps) | -| --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_NV6ads_A10_v5 |6 |55 |180 | 1/6 | 4 | 4 | 6400 / 100 | 2 / 5000 | -| Standard_NV12ads_A10_v5 |12 |110 |360 | 1/3 | 8 | 4 | 12800 / 200 | 2 / 10000 | -| Standard_NV18ads_A10_v5 |18 |220 |720 | 1/2 | 12 | 8 | 25600 / 384 | 4 / 20000 | -| Standard_NV36ads_A10_v5 |36 |440 |1440 | 1 | 24 | 16 |51200 / 768 | 4 / 40000 | -| Standard_NV36adms_A10_v5 |36 |880 |2880 | 1 | 24 | 32 |51200 / 768 | 8 / 80000 | -| Standard_NV72ads_A10_v5 |72 |880 |2880 | 2 | 48 | 32 | 80000 / 1200 | 8 / 80000 | - -1 NVadsA10v5-series VMs feature AMD Simultaneous multithreading Technology - -2 The actual GPU VRAM reported in the operating system will be a little less due to Error Correcting Code (ECC) support. - - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure NVadsA10v5-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nvv3-series.md b/articles/virtual-machines/nvv3-series.md deleted file mode 100644 index 98dfa6a7df2..00000000000 --- a/articles/virtual-machines/nvv3-series.md +++ /dev/null @@ -1,66 +0,0 @@ ---- -title: NVv3-series - Azure Virtual Machines -description: Specifications for the NVv3-series VMs. -services: virtual-machines -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 12/21/2022 -ms.author: vikancha ---- - -# NVv3-series - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NVv3-series virtual machines are powered by [NVIDIA Tesla M60](https://images.nvidia.com/content/tesla/pdf/188417-Tesla-M60-DS-A4-fnl-Web.pdf) GPUs and NVIDIA GRID technology with Intel E5-2690 v4 (Broadwell) CPUs and Intel Hyper-Threading Technology. These virtual machines are targeted for GPU accelerated graphics applications and virtual desktops where customers want to visualize their data, simulate results to view, work on CAD, or render and stream content. Additionally, these virtual machines can run single precision workloads such as encoding and rendering. NVv3 virtual machines support Premium Storage and come with twice the system memory (RAM) when compared with its predecessor NV-series. - -Each GPU in NVv3 instances comes with a GRID license. This license gives you the flexibility to use an NV instance as a virtual workstation for a single user, or 25 concurrent users can connect to the VM for a virtual application scenario. - -[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs / Expected network bandwidth (Mbps) | Virtual Workstations | Virtual Applications | -|---|---|---|---|---|---|---|---|---|---|---| -| Standard_NV12s_v3 |12 | 112 | 320 | 1 | 8 | 12 | 20000/200 | 4 / 6000 | 1 | 25 | -| Standard_NV24s_v3 |24 | 224 | 640 | 2 | 16 | 24 | 40000/400 | 8 / 12000 | 2 | 50 | -| Standard_NV48s_v3 |48 | 448 | 1280 | 4 | 32 | 32 | 80000/800 | 8 / 24000 | 4 | 100 | - -1 1 GPU = one-half M60 card. - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure N-series VMs, NVIDIA GPU drivers must be installed. - -The [NVIDIA GPU Driver Extension](./extensions/hpccompute-gpu-windows.md) installs appropriate NVIDIA CUDA or GRID drivers on an N-series VM. Install or manage the extension using the Azure portal or tools such as Azure PowerShell or Azure Resource Manager templates. See the [NVIDIA GPU Driver Extension documentation](./extensions/hpccompute-gpu-windows.md) for supported operating systems and deployment steps. For general information about VM extensions, see [Azure virtual machine extensions and features](./extensions/overview.md). - -If you choose to install NVIDIA GPU drivers manually, see [N-series GPU driver setup for Windows](./windows/n-series-driver-setup.md) or [N-series GPU driver setup for Linux](./linux/n-series-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/nvv4-series.md b/articles/virtual-machines/nvv4-series.md deleted file mode 100644 index 353f659b831..00000000000 --- a/articles/virtual-machines/nvv4-series.md +++ /dev/null @@ -1,67 +0,0 @@ ---- -title: NVv4-series -description: Specifications for the NVv4-series VMs. -author: vikancha-MSFT -ms.service: azure-virtual-machines -ms.subservice: sizes -ms.topic: conceptual -ms.date: 01/12/2020 -ms.author: vikancha ---- - -# NVv4-series - -**Applies to:** :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The NVv4-series virtual machines are powered by [AMD Radeon Instinct MI25](https://www.amd.com/en/newsroom/press-releases/2020-3-25-2nd-gen-amd-epyc-processors-and-amd-radeon-instin.html) GPUs and AMD EPYC 7V12(Rome) CPUs with a base frequency of 2.45GHz, all-cores peak frequency of 3.1GHz and single-core peak frequency of 3.3GHz. With NVv4-series Azure is introducing virtual machines with partial GPUs. Pick the right sized virtual machine for GPU accelerated graphics applications and virtual desktops starting at 1/8th of a GPU with 2 GiB frame buffer to a full GPU with 16 GiB frame buffer. NVv4 virtual machines currently support only Windows guest operating system. - -
- -[ACU](acu.md): 230-260
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Supported
-[Ultra Disks](disks-types.md#ultra-disks): Supported ([Learn more](https://techcommunity.microsoft.com/t5/azure-compute/ultra-disk-storage-for-hpc-and-gpu-vms/ba-p/2189312) about availability, usage and performance)
-[Live Migration](maintenance-and-updates.md): Not Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Not Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | GPU | GPU memory: GiB | Max data disks | Max uncached disk throughput: IOPS/MBps | Max NICs / Expected network bandwidth (MBps) | -| --- | --- | --- | --- | --- | --- | --- | --- | --- | -| Standard_NV4as_v4 |4 |14 |88 | 1/8 | 2 | 4 | 6400 / 96 | 2 / 1000 | -| Standard_NV8as_v4 |8 |28 |176 | 1/4 | 4 | 8 | 12800 / 192 | 4 / 2000 | -| Standard_NV16as_v4 |16 |56 |352 | 1/2 | 8 | 16 | 25600 / 384 | 8 / 4000 | -| Standard_NV32as_v4 |32 |112 |704 | 1 | 16 | 32 | 51200 / 768 |8 / 8000 | - -1 NVv4-series VMs feature AMD Simultaneous multithreading Technology - - - -## Supported operating systems and drivers - -To take advantage of the GPU capabilities of Azure NVv4-series VMs running Windows, AMD GPU drivers must be installed. - -To install AMD GPU drivers manually, see [N-series AMD GPU driver setup for Windows](./windows/n-series-amd-driver-setup.md) for supported operating systems, drivers, installation, and verification steps. - - -[!INCLUDE [virtual-machines-common-sizes-table-defs](./includes/virtual-machines-common-sizes-table-defs.md)] - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) -- [Previous generations](sizes-previous-gen.md) - -Pricing Calculator : [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -For more information on disk types, see [What disk types are available in Azure?](disks-types.md) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. diff --git a/articles/virtual-machines/sizes-b-series-burstable.md b/articles/virtual-machines/sizes-b-series-burstable.md deleted file mode 100644 index e2dbd1432bb..00000000000 --- a/articles/virtual-machines/sizes-b-series-burstable.md +++ /dev/null @@ -1,69 +0,0 @@ ---- -title: B-series burstable - Azure Virtual Machines -description: Describes the B-series of burstable Azure VM sizes. -services: virtual-machines -ms.subservice: sizes -author: rishabv90 -ms.service: azure-virtual-machines -ms.topic: conceptual -ms.date: 02/03/2020 -ms.author: risverma ---- - -# B-series burstable virtual machine sizes - -**Applies to:** :heavy_check_mark: Linux VMs :heavy_check_mark: Windows VMs :heavy_check_mark: Flexible scale sets :heavy_check_mark: Uniform scale sets - -The B-series VMs can be deployed on a variety of hardware types and processors, so competitive bandwidth allocation is provided. B-series run on the 3rd Generation Intel® Xeon® Platinum 8370C (Ice Lake), the Intel® Xeon® Platinum 8272CL (Cascade Lake), the Intel® Xeon® 8171M 2.1 GHz (Skylake), the Intel® Xeon® E5-2673 v4 2.3 GHz (Broadwell), or the Intel® Xeon® E5-2673 v3 2.4 GHz (Haswell) processors. B-series VMs are ideal for workloads that do not need the full performance of the CPU continuously, like web servers, proof of concepts, small databases and development build environments. These workloads typically have burstable performance requirements. To determine the physical hardware on which this size is deployed, query the virtual hardware from within the virtual machine. The B-series provides you with the ability to purchase a VM size with baseline performance that can build up credits when it is using less than its baseline. When the VM has accumulated credits, the VM can burst above the baseline using up to 100% of the vCPU when your application requires higher CPU performance. - -The B-series comes in the following VM sizes: - -[Azure Compute Unit (ACU)](./acu.md): Varies*
-[Premium Storage](premium-storage-performance.md): Supported
-[Premium Storage caching](premium-storage-performance.md): Not Supported
-[Live Migration](maintenance-and-updates.md): Supported
-[Memory Preserving Updates](maintenance-and-updates.md): Supported
-[VM Generation Support](generation-2.md): Generation 1 and 2
-[Accelerated Networking](/azure/virtual-network/create-vm-accelerated-networking-cli): Supported**
-[Ephemeral OS Disks](ephemeral-os-disks.md): Supported
-[Nested Virtualization](/virtualization/hyper-v-on-windows/user-guide/nested-virtualization): Not Supported
- -*B-series VMs are burstable and thus ACU numbers will vary depending on workloads and core usage.
-**Accelerated Networking is only supported for *Standard_B12ms*, *Standard_B16ms* and *Standard_B20ms*. -
-
- -| Size | vCPU | Memory: GiB | Temp storage (SSD) GiB | Base CPU Performance of VM (%) | Initial Credits | Credits banked/hour | Max Banked Credits | Max data disks | Max uncached disk throughput: IOPS/MBps | Max burst uncached disk throughput: IOPS/MBps1 | Max NICs | -|----------------|------|-------------|------------------------|--------------------------------|-----------------|---------------------|--------------------|----------------|-----------------------------------------|------------------------------------------------|----------| -| Standard_B1ls2 | 1 | 0.5 | 4 | 5 | 30 | 3 | 72 | 2 | 160/10 | 4000/100 | 2 | -| Standard_B1s | 1 | 1 | 4 | 10 | 30 | 6 | 144 | 2 | 320/10 | 4000/100 | 2 | -| Standard_B1ms | 1 | 2 | 4 | 20 | 30 | 12 | 288 | 2 | 640/10 | 4000/100 | 2 | -| Standard_B2s | 2 | 4 | 8 | 20 | 60 | 24 | 576 | 4 | 1280/15 | 4000/100 | 3 | -| Standard_B2ms | 2 | 8 | 16 | 30 | 60 | 36 | 864 | 4 | 1920/22.5 | 4000/100 | 3 | -| Standard_B4ms | 4 | 16 | 32 | 22.5 | 120 | 54 | 1296 | 8 | 2880/35 | 8000/200 | 4 | -| Standard_B8ms | 8 | 32 | 64 | 17 | 240 | 81 | 1994 | 16 | 4320/50 | 8000/200 | 4 | -| Standard_B12ms | 12 | 48 | 96 | 17 | 360 | 121 | 2908 | 16 | 4320/50 | 16000/400 | 6 | -| Standard_B16ms | 16 | 64 | 128 | 17 | 480 | 162 | 3888 | 32 | 4320/50 | 16000/400 | 8 | -| Standard_B20ms | 20 | 80 | 160 | 17 | 600 | 202 | 4867 | 32 | 4320/50 | 16000/400 | 8 | - -1 B-series VMs can [burst](./disk-bursting.md) their disk performance and get up to their bursting max for up to 30 minutes at a time. - -2 B1ls is supported only on Linux - - -## Other sizes and information - -- [General purpose](sizes-general.md) -- [Compute optimized](sizes-compute.md) -- [Memory optimized](sizes-memory.md) -- [Storage optimized](sizes-storage.md) -- [GPU optimized](sizes-gpu.md) -- [High performance compute](sizes-hpc.md) - -Pricing Calculator: [Pricing Calculator](https://azure.microsoft.com/pricing/calculator/) - -More information on Disks Types: [Disk Types](./disks-types.md#ultra-disks) - -## Next steps - -Learn more about how [Azure compute units (ACU)](acu.md) can help you compare compute performance across Azure SKUs. From c6d9125b048542d32299f4373ff044dcff1a69ff Mon Sep 17 00:00:00 2001 From: Matt McInnes Date: Fri, 6 Sep 2024 14:38:30 -0700 Subject: [PATCH 3/4] Fixing redirect file names --- ...enpublishing.redirection.virtual-machines.json | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-) diff --git a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json index 01d35f5f1e5..a1b784b6e05 100644 --- a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json +++ b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json @@ -6397,7 +6397,7 @@ }, { "source_path_from_root": "/articles/virtual-machines/nva10v5-series.md", - "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nva10v5-series", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nvadsa10v5-series", "redirect_document_id": true }, { @@ -6407,7 +6407,7 @@ }, { "source_path_from_root": "/articles/virtual-machines/np-series.md", - "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/np-series", + "redirect_url": "/azure/virtual-machines/sizes/fpga-accelerated/np-series", "redirect_document_id": true }, { @@ -6427,7 +6427,7 @@ }, { "source_path_from_root": "/articles/virtual-machines/nda100-v4-series.md", - "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nda100v4-series", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ndasra100v4-series", "redirect_document_id": true }, { @@ -6440,11 +6440,6 @@ "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ndh100v5-series", "redirect_document_id": true }, - { - "source_path_from_root": "/articles/virtual-machines/nvv3-series.md", - "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nvv3-series", - "redirect_document_id": true - }, { "source_path_from_root": "/articles/virtual-machines/ncv2-series.md", "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncv2-series", @@ -6452,7 +6447,7 @@ }, { "source_path_from_root": "/articles/virtual-machines/nct4-v3-series.md", - "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/nct4v3-series", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncast4v3-series", "redirect_document_id": true }, { @@ -6467,7 +6462,7 @@ }, { "source_path_from_root": "/articles/virtual-machines/hx-series.md", - "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hx-series.", + "redirect_url": "/azure/virtual-machines/sizes/high-performance-compute/hx-series", "redirect_document_id": true }, { From 72c69e923c5f5b548ef47ed12a60a499a78a7d01 Mon Sep 17 00:00:00 2001 From: Matt McInnes Date: Fri, 6 Sep 2024 14:42:17 -0700 Subject: [PATCH 4/4] Fixing redirect file names --- .../.openpublishing.redirection.virtual-machines.json | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json index a1b784b6e05..b18039578db 100644 --- a/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json +++ b/articles/virtual-machines/.openpublishing.redirection.virtual-machines.json @@ -6445,6 +6445,11 @@ "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncv2-series", "redirect_document_id": true }, + { + "source_path_from_root": "/articles/virtual-machines/ncv3-series.md", + "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncv3-series", + "redirect_document_id": true + }, { "source_path_from_root": "/articles/virtual-machines/nct4-v3-series.md", "redirect_url": "/azure/virtual-machines/sizes/gpu-accelerated/ncast4v3-series",