hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Liang Xie (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-5776) Support 'hedged' reads in DFSClient
Date Thu, 20 Feb 2014 05:02:26 GMT

    [ https://issues.apache.org/jira/browse/HDFS-5776?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13906611#comment-13906611
] 

Liang Xie commented on HDFS-5776:
---------------------------------

Huge thanks, [~saint.ack@gmail.com] !  Here is the YCSB test result w/ the latest v21 patch:
My test env, setting, data just same as the previous ones.

hedged read thread pool size = 0:
{code}
[OVERALL], Throughput(ops/sec), 248.69758063844773
[READ], AverageLatency(us), 40152.776944654586
[READ], 50thPercentileLatency(us), 21023
[READ], 95thPercentileLatency(us), 151130
[READ], 99thPercentileLatency(us), 255973
[READ], 99.9thPercentileLatency(us), 420971  
{code}

hedged read thread pool size = 50, hedged read timeout = 250ms :
{code}
[OVERALL], Throughput(ops/sec), 231.90953643421446 
[READ], AverageLatency(us), 43094.41323982726 
[READ], 50thPercentileLatency(us), 23280
[READ], 95thPercentileLatency(us), 160024
[READ], 99thPercentileLatency(us), 258110
[READ], 99.9thPercentileLatency(us), 278328
{code}

hedged read thread pool size = 50, hedged read timeout = 150ms :
{code}
[OVERALL], Throughput(ops/sec), 240.76033437917079
[READ], AverageLatency(us), 41507.364070480435
[READ], 50thPercentileLatency(us), 23316
[READ], 95thPercentileLatency(us), 158907
[READ], 99thPercentileLatency(us), 168504
[READ], 99.9thPercentileLatency(us), 218926
{code}

hedged read thread pool size = 50, hedged read timeout = 100ms :
{code}
[OVERALL], Throughput(ops/sec), 271.0941444451295
[READ], AverageLatency(us), 36863.66854547243
[READ], 50thPercentileLatency(us), 21371
[READ], 95thPercentileLatency(us), 114943
[READ], 99thPercentileLatency(us), 121673
[READ], 99.9thPercentileLatency(us), 195467
{code}

> Support 'hedged' reads in DFSClient
> -----------------------------------
>
>                 Key: HDFS-5776
>                 URL: https://issues.apache.org/jira/browse/HDFS-5776
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: hdfs-client
>    Affects Versions: 3.0.0
>            Reporter: Liang Xie
>            Assignee: Liang Xie
>         Attachments: HDFS-5776-v10.txt, HDFS-5776-v11.txt, HDFS-5776-v12.txt, HDFS-5776-v12.txt,
HDFS-5776-v13.wip.txt, HDFS-5776-v14.txt, HDFS-5776-v15.txt, HDFS-5776-v17.txt, HDFS-5776-v17.txt,
HDFS-5776-v2.txt, HDFS-5776-v3.txt, HDFS-5776-v4.txt, HDFS-5776-v5.txt, HDFS-5776-v6.txt,
HDFS-5776-v7.txt, HDFS-5776-v8.txt, HDFS-5776-v9.txt, HDFS-5776.txt, HDFS-5776v18.txt, HDFS-5776v21.txt
>
>
> This is a placeholder of hdfs related stuff backport from https://issues.apache.org/jira/browse/HBASE-7509
> The quorum read ability should be helpful especially to optimize read outliers
> we can utilize "dfs.dfsclient.quorum.read.threshold.millis" & "dfs.dfsclient.quorum.read.threadpool.size"
to enable/disable the hedged read ability from client side(e.g. HBase), and by using DFSQuorumReadMetrics,
we could export the interested metric valus into client system(e.g. HBase's regionserver metric).
> The core logic is in pread code path, we decide to goto the original fetchBlockByteRange
or the new introduced fetchBlockByteRangeSpeculative per the above config items.



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message