[jira] [Commented] (HBASE-20312) CCSMap: A faster, GC-friendly, less memory Concurrent Map for memstore

["Ted Yu (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-20312?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16422077#comment-16422077 ] 

Ted Yu commented on HBASE-20312:
--------------------------------

Excited to see the master patch.
Thanks for the hard work, Chance.

hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/skiplist/core/IAllocatorHandler.java

To my knowledge, there is no need to prefix class name with 'I' - IAllocatorHandler is an interface. I think AllocatorHandler should suffice.

One question I have is: does it make sense to give individual table the ability to switch CCSMap on / off ? The thinking is that different tables would have varying schema where the degree of benefit of CCSMap varies.

Thanks

> CCSMap: A faster, GC-friendly, less memory Concurrent Map for memstore
> ----------------------------------------------------------------------
>
>                 Key: HBASE-20312
>                 URL: https://issues.apache.org/jira/browse/HBASE-20312
>             Project: HBase
>          Issue Type: New Feature
>          Components: regionserver
>            Reporter: Xiang Wang
>            Assignee: Chance Li
>            Priority: Minor
>             Fix For: 3.0.0
>
>         Attachments: HBASE-20312-1.3.2.patch, HBASE-20312-master.v1.patch, ccsmap-branch-1.1.patch, jira1.png, jira2.png, jira3.png
>
>
> Now hbase use ConcurrentSkipListMap as memstore's data structure.
> Although MemSLAB reduces memory fragment brought by key-value pairs.
> Hundred of millions key-value pairs still make young generation garbage-collection(gc) stop time long.
>  
> These are 2 gc problems of ConcurrentSkipListMap:
> 1. HBase needs 3 objects to store one key-value on expectation. One Index(skiplist's average node height is 1), one Node, and one KeyValue. Too many objects are created for memstore.
> 2. Recent inserted KeyValue and its map structure(Index, Node) are assigned on young generation.The card table (for CMS gc algorithm) or RSet(for G1 gc algorithm) will change frequently on high writing throughput, which makes YGC slow.
>  
> We devleoped a new skip-list map called CompactdConcurrentSkipListMap(CCSMap for short),
> which provides similary features like ConcurrentSkipListMap but get rid of Objects for every key-value pairs.
> CCSMap's memory structure is like this picture:
> !jira1.png!
>  
> One CCSMap consists of a certain number of Chunks. One Chunk consists of a certain number of nodes. One node is corresspding one element. This element's all information and its key-value is encoded on a continuous memory segment without any objects.
> Features:
> 1. all insert,update,delete operations is lock-free on CCSMap.
> 2. Consume less memory, it brings 40% memory saving for 50Byte key-value.
> 3. Faster on small key-value because of better cacheline usage. 20~30% better read/write troughput than ConcurrentSkipListMap for 50Byte key-value.
> CCSMap do not support recyle space when deleting element. But it doesn't matter for hbase because of region flush.
> CCSMap has been running on Alibaba's hbase clusters over 17 months, it cuts down YGC time significantly. here are 2 graph of before and after.
> !jira2.png!
> !jira3.png!
>  
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)