utsaslab / pebblesdb

The PebblesDB write-optimized key-value store (SOSP 17)
BSD 3-Clause "New" or "Revised" License
506 stars 99 forks source link

Extend PebblesDB to use SURF filters #17

Open vijay03 opened 5 years ago

vijay03 commented 5 years ago

PebblesDB currently uses bloom filters. Change PebblesDB so that it uses SURF Filters from CMU (https://github.com/efficient/SuRF) instead. This should help both point-query and range query performance.

malavdesai16 commented 3 years ago

Hello Sir, It was noted in the paper that using Bloom filters is PebblesDB leads to high consumption of memory compared to HyperLevelDB. I read an interesting paper on Vacuum Filters. Is it feasible to extend PebblesDB with Vacuum Filters? Although, it wont help point-query and range query performance but it will help with minimizing the space consumption.