You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-dev@hadoop.apache.org by "Allen Wittenauer (JIRA)" <ji...@apache.org> on 2014/07/18 00:08:06 UTC
[jira] [Resolved] (HDFS-296) Serial streaming performance should be
Math.min(ideal client performance, ideal serial hdfs performance)
[ https://issues.apache.org/jira/browse/HDFS-296?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Allen Wittenauer resolved HDFS-296.
-----------------------------------
Resolution: Incomplete
I'm going to close this as stale given how many changes have happened to both streaming and HDFS since this was filed.
> Serial streaming performance should be Math.min(ideal client performance, ideal serial hdfs performance)
> --------------------------------------------------------------------------------------------------------
>
> Key: HDFS-296
> URL: https://issues.apache.org/jira/browse/HDFS-296
> Project: Hadoop HDFS
> Issue Type: Improvement
> Environment: Mac OS X 10.5.2, Java 6
> Reporter: Sam Pullara
>
> I looked at all the code long and hard and this was my analysis (could be wrong, I'm not an expert on this codebase):
> Current Serial HDFS performance = Average Datanode Performance
> Average Datanode Performance = Average Disk Performance (even if you have more than one)
> We should have:
> Ideal Serial HDFS Performance = Sum of Ideal Datanode Performance
> Ideal Datanode Performance = Sum of disk performance
> When you read a single file serially from HDFS there are a number of limitations that come into play:
> 1) Blocks on multiple datanodes will be load balanced between them - averaging the performance of the datanodes
> 2) Blocks on multiple disks in a single datanode are load balanced between them - averaging the performance of the disks
> I think that all this could be fixed if we actually prefetched fully read blocks on the client until the client can no longer keep up with the data or there is another bottleneck like network bandwidth.
> This seems like a reasonably common use case though not the typical MapReduce case.
--
This message was sent by Atlassian JIRA
(v6.2#6252)