Skip to content

Latest commit

 

History

History
57 lines (51 loc) · 2.47 KB

description.md

File metadata and controls

57 lines (51 loc) · 2.47 KB
Raw

Distributed Cache

This is project description and list of TODOs.

Completed tasks
  • Implement LRU eviction policy for cache
  • Adding a server has two steps:
    • Start the server.
    • Ask the cache server to spwan. Then it will listen for new connections.
  • Implement get, set, delete
  • Handle timeout
  • Log important stats like hits and misses
  • Implement consistent hashing
  • Deal with unavailable servers like cache miss. When they come back online again, they will be treated as new servers.
  • Implement health checks
  • (De)Serialization of key cache operations so that it can be reconstructed in the event of crash
  • Decouple cache design from cache client. Allows you to swap the cache design on the fly. And opens up cache design for testing.
  • Restructure the project so that files can be imported properly.
  • Add setup.py and other information for pypi.
  • Write documentation
  • Write in batches to the disk. Make provision for flush as well.
  • Implement increment and decrement in redis like counter
  • Write binary objects (image, pdfs, etc) to the disk
  • Multiple methods of cache reconstruction and persistence.
TODO for Release/General
  • Make some calls aysnc after testing code on single thread
  • Test everything!
    • Test all the different types of objects supported
    • Test cache
    • Test consistent hashing. Make it more uniform
    • Test heartbeat/health check
    • Test logger
    • Test communication (maybe the hardest part)
    • Test serialization (and probably replace it with a faster library)
  • Add benchmarks for databases
    • Create 100,000 files of products. Read and query these files.
    • Create 100,000 files of random images maybe (binary). Read and query these files.
    • Note: For benchmarks, we consider reading from database/file directly versus having querying in-memory key, value store.
  • Add lots of sample code
  • Write wiki
TODO for communication
  • CacheServer starts HealthServers. CacheClients start HealthClients.
  • Make sure it works for single server node i.e. the communication between client and server works flawlessly. Cache is fine. :)
  • Allow users to change the communication protocol: TCP/UDP/Unix Sockets.
TODO for persistent storage
  • When key is not in the cache, check for it in the disk/database
TODO for Cache Design
  • Expiration time on keys. It just makes each record expensive to store.
  • Implement different types of cache eviction strategies like LFU, etc.
  • Determine the size of the cache being used.