Node Utilization

The primary level node monitoring graph includes CPU and Memory utilization graphs superimposed in the same chart which can be seen on Node Details page as shown below.

![] (https://raw.githubusercontent.com/wiki/impetus-opensource/ankush/images/Image102.png) Figure 102: Utilization Graph of CPU/Memory

Detailed node level monitoring graphs are accessible by clicking “>” in Utilization Graph section which opens up node level performance graphs.

Monitoring Performance Graph – Node Level

This includes various System Metrics graphs (i.e. CPU, Memory, Load, Network, and Packet) and other technology specific graphs organized across various relevant categories.

Note:

1. The technology specific graphs are covered in Node Utilization section of individual Technology specific guide.

2. System Metrics graphs are independent of technology.

The graphs are displayed in a user-friendly tree view based categorization structure as shown below:

![] (https://raw.githubusercontent.com/wiki/impetus-opensource/ankush/images/Image103.png)

Figure 103: Tree level graphs - Node level

The graphs are viewport context based enabled. This ensures that even the selected graphs load only when the user scrolls through it. This sets the context and enhances the performance of the application. Once the user scrolls through the graphs, each graph swiftly appears on to the screen for reference monitoring.

![] (https://raw.githubusercontent.com/wiki/impetus-opensource/ankush/images/Image104.png) Figure 104: Utilization Metrics

All the graphs are not pre-fetched in one go for display. In order to enhance the performance of the application and to utilize the resources like bandwidth smartly, the graphs are loaded on demand as user scrolls down.

Note: How to differentiate between performance graphs at cluster and node level?

The heading of graphs at cluster level displays the cluster name while the heading of graphs or page at node level displays the IP of the node.

![] (https://raw.githubusercontent.com/wiki/impetus-opensource/ankush/images/Image105.png) Figure 105: Performance graphs cluster / node level