Efficient way to split a DB into 2 or multiple DB based on key range?
Open
design discussion
- Dominant language
- C++
- Stars
- 32.1k
- Forks
- 6.9k
- Avg merge
- 32m
- Merged PRs (30d)
- 1
Description
Is there an existing function or logic to split one DB into 2 or more DB based on the key range?
For example, the current DB key range is [0, 1000]. To split the DB into 2 small DBs with a balanced space size, the 2 new DB may be [0, 420] and [430, 1000] (say each contains 200 KV-pairs).
Theoretically, we just need to split at most 1 SST file in each level if it is a 2-DB split. Other SST files can be a trivial move.
Contributor guide
Assessment
This issue has not been assessed yet.