[
https://issues.apache.org/jira/browse/HADOOP-3288?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12591233#action_12591233
]
Sam Pullara commented on HADOOP-3288:
-------------------------------------
This improvement is targeted specifically at a single client reading from (or
writing to) an HDFS cluster as fast as possible.
1) RAID of the nodes in question would typically fix that case. But if thats
the solution it isn't clear to me why we have the option of multiple disks per
DataNode.
2) A single client only. For example, copying a very large result file from
HDFS. Some examples include very large (1 TB) images produced for printing (HP
use case) and storage of VM images (another use case).
> Serial streaming performance should be Math.min(ideal client performance,
> ideal serial hdfs performance)
> --------------------------------------------------------------------------------------------------------
>
> Key: HADOOP-3288
> URL: https://issues.apache.org/jira/browse/HADOOP-3288
> Project: Hadoop Core
> Issue Type: Improvement
> Components: dfs
> Affects Versions: 0.16.3, 0.18.0
> Environment: Mac OS X 10.5.2, Java 6
> Reporter: Sam Pullara
> Fix For: 0.18.0
>
>
> 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 is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
