Danil Lipovoy created HBASE-23887:
-------------------------------------
Summary: BlockCache performance improve
Key: HBASE-23887
URL: https://issues.apache.org/jira/browse/HBASE-23887
Project: HBase
Issue Type: New Feature
Reporter: Danil Lipovoy
Attachments: 2jira.png
Hi!
I first time here, sorry if something wrong.
I want propose how to improve performance when data in HFiles much more than
BlockChache (usual story in BigData). The idea - caching only part of DATA
blocks. It is good becouse LruBlockCache starts to work and save huge amount of
GC. See the picture in attachment with test below. Requests per second is
higher, GC is lower.
The key point of the code:
Added the parameter: *hbase.lru.cache.data.block.percent* which by default = 100
But if we set it 0-99, then will work the next logic:
*public void* cacheBlock(BlockCacheKey cacheKey, Cacheable buf, *boolean*
inMemory) {
*if* (_cacheDataBlockPercent_ != 100 && buf.getBlockType().isData()) {
*if* (cacheKey.getOffset() % 100 >= _cacheDataBlockPercent_)
*return*;
}
...
}
Descriptions of the test:
3 nodes E5-2698 v4 @ 2.20GHz, 700 Gb Mem.
3 RegionServers
6 tables by 64 regions by 1.88 Gb data in each = 720 Gb total (only FAST_DIFF)
Total BlockCache Size = 38 Gb
Random read in 24 threads
I am going to make Pull Request, hope it is right way to make some contribution
in this cool product. Correct me please if something wrong.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
