[jira] [Issue Comment Deleted] (HBASE-21657) PrivateCellUtil#estimatedSerializedSizeOf has been the bottleneck in 100% scan case.

["Zheng Hu (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HBASE-21657?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Zheng Hu updated HBASE-21657:
-----------------------------
    Comment: was deleted

(was: In HFileReaderImpl#getCell,   we will new an SizeCachedKeyValue for each cell,  but it only cache rowLen & keyLen now, I think we can add a serializedSize in here which may save lots of cpu.)

> PrivateCellUtil#estimatedSerializedSizeOf has been the bottleneck in 100% scan case.
> ------------------------------------------------------------------------------------
>
>                 Key: HBASE-21657
>                 URL: https://issues.apache.org/jira/browse/HBASE-21657
>             Project: HBase
>          Issue Type: Bug
>          Components: Performance
>            Reporter: Zheng Hu
>            Assignee: Zheng Hu
>            Priority: Major
>             Fix For: 3.0.0, 2.2.0, 2.1.3, 2.0.5
>
>         Attachments: hbase20-ssd-100-scan-traces.svg
>
>
> We are evaluating the performance of branch-2, and find that the throughput of scan in SSD cluster is almost the same as HDD cluster. so I made a FlameGraph on RS, and found that the PrivateCellUtil#estimatedSerializedSizeOf cost about 29% cpu, Obviously, it has been the bottleneck in 100% scan case.
> See the [^hbase20-ssd-100-scan-traces.svg]
> BTW, in our XiaoMi branch, we introduce a HRegion#updateReadRequestsByCapacityUnitPerSecond to sum up the size of cells (for metric monitor), so it seems the performance loss was amplified.



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