[
https://issues.apache.org/jira/browse/HBASE-16213?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15414747#comment-15414747
]
Anoop Sam John commented on HBASE-16213:
----------------------------------------
Thanks for the detailed test and info. The seek perf improvement is
excellent. The size delta because of row offset meta data seems very small
when cell size is not so small (>100 bytes I consider) We need this info so
that the doc around this feature can explain it well how much block size
getting increased because of this new type of DBE. Small suggestion in your
ppt 1st page, just make one more column and explicitly add meta data size.
Else every time read, we have to subtract 2 values. And may be add a field
like overhead % also. (ie. (Total size - data size)/data size)
> A new HFileBlock structure for fast random get
> ----------------------------------------------
>
> Key: HBASE-16213
> URL: https://issues.apache.org/jira/browse/HBASE-16213
> Project: HBase
> Issue Type: New Feature
> Components: Performance
> Reporter: binlijin
> Assignee: binlijin
> Attachments: HBASE-16213-master_v1.patch, HBASE-16213.patch,
> HBASE-16213_branch1_v3.patch, HBASE-16213_v2.patch, hfile-cpu.png,
> hfile_block_performance.pptx, new-hfile-block.xlsx
>
>
> HFileBlock store cells sequential, current when to get a row from the block,
> it scan from the first cell until the row's cell.
> The new structure store every row's start offset with data, so it can find
> the exact row with binarySearch.
> I use EncodedSeekPerformanceTest test the performance.
> First use ycsb write 100w data, every row have only one qualifier, and
> valueLength=16B/64/256B/1k.
> Then use EncodedSeekPerformanceTest to test random read 1w or 100w row, and
> also record HFileBlock's dataSize/dataWithMetaSize in the encoding.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
