[
https://issues.apache.org/jira/browse/HDFS-7847?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15012903#comment-15012903
]
Colin Patrick McCabe commented on HDFS-7847:
--------------------------------------------
bq. If understand this correctly, you are not targeting to build a benchmark,
but rather unit tests of NameNode RPC. Isn't there a bunch of those in HDFS
source tree already? Just trying to understand what is different.
Excuse me, I mis-spoke. {{NNThroughputBenchmark}} is a benchmark, not a unit
test.
bq. Have you looked at org.apache.hadoop.fs.loadGenerator.LoadGenerator. Would
its functionality suffice the testing needs you have in mind?
{{LoadGenerator}} is certainly much more similar to {{NNThroughputBenchmark}}
than any of the other things you mentioned. However, it performs a mix of only
list, read, and write operations. {{NNThroughputBenchmark}} performs a wider
range of operations. In general, I would like to avoid read and write
operations in my testing of NN RPC performance, since those operations involve
the DataNode heavily. I think it's good to have multiple different benchmarks
testing different things.
> Modify NNThroughputBenchmark to be able to operate on a remote NameNode
> -----------------------------------------------------------------------
>
> Key: HDFS-7847
> URL: https://issues.apache.org/jira/browse/HDFS-7847
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 2.8.0
> Reporter: Colin Patrick McCabe
> Assignee: Charles Lamb
> Fix For: 2.8.0
>
> Attachments: HDFS-7847.000.patch, HDFS-7847.001.patch,
> HDFS-7847.002.patch, HDFS-7847.003.patch, HDFS-7847.004.patch,
> HDFS-7847.005.patch, make_blocks.tar.gz
>
>
> Modify NNThroughputBenchmark to be able to operate on a NN that is not in
> process. A followon Jira will modify it some more to allow quantifying native
> and java heap sizes, and some latency numbers.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
