RocksDB Advanced Features and Optimizations - Lightning Talks

RocksDB Advanced Features and Optimizations - Lightning Talks

Meta Developers via YouTube Direct link

Future work.

17 of 40

17 of 40

Future work.

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

RocksDB Advanced Features and Optimizations - Lightning Talks

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Recap: Compaction A RocksDB operation that merges SST files
  3. 3 Recap: Compaction Filter A customizable logic to delete key-value pairs during compaction
  4. 4 Lua Compaction Filter You can write Compaction Filters in Lua!
  5. 5 How to Use?
  6. 6 CPU Overhead CPU usage in one Facebook internal use case
  7. 7 Limitations
  8. 8 Time-series data
  9. 9 Writing a key
  10. 10 Insert into memtable
  11. 11 Reduce CPU usage for binary search
  12. 12 Benchmarks
  13. 13 Key Benefit
  14. 14 Usage Open the DB
  15. 15 Design Considerations
  16. 16 Benchmark - (WA)
  17. 17 Future work.
  18. 18 What is SeekForPrev?
  19. 19 Motivation
  20. 20 Basic Logic What you do previously to achieve same functionality
  21. 21 Implementation sketch
  22. 22 Direct 1/0
  23. 23 Write Path - Memtable
  24. 24 Write Path - Flush
  25. 25 Write Path - Compaction
  26. 26 Bulkload-Create SST file
  27. 27 Bulkloading - Level criteria
  28. 28 Bulkload - memtable conflict
  29. 29 Use cases
  30. 30 Shard Migration DB1
  31. 31 Import from Hadoop
  32. 32 Bulkload - Create SST file
  33. 33 Bulkload - Ingest SST file
  34. 34 Cloning a DB
  35. 35 Before WDT...
  36. 36 Facebook internal usage of WDT (cont.) Bytes per minute per use case
  37. 37 Java 8 Streams
  38. 38 Secret Sauce?
  39. 39 Iterlib::dynamic
  40. 40 Iterators

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.