[
https://issues.apache.org/jira/browse/HDFS-3696?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Tsz Wo (Nicholas), SZE updated HDFS-3696:
-----------------------------------------
Attachment: h3696_20120724.patch
h3696_20120724.patch: add setChunkedStreamingMode(32kB).
I tried several chunk sizes for writing 300MB files. 32kB was the best in my
test.
|| Chunk size || 1st || 2nd ||
| 4kB | 3.95MB/s | 3.95MB/s |
| 16kB | 7.81MB/s | 7.70MB/s |
| 24kB | 12.58MB/s | 12.29MB/s |
| 32kB | 14.15MB/s | 14.28MB/s |
| 48kB | 14.25MB/s | 13.29MB/s |
| 64kB | 13.65MB/s | 13.57MB/s |
| 128kB | 13.94MB/s | 13.15MB/s |
| 1MB | 13.11MB/s | 13.45MB/s |
> FsShell put using WebHdfsFileSystem goes OOM when file size is big
> ------------------------------------------------------------------
>
> Key: HDFS-3696
> URL: https://issues.apache.org/jira/browse/HDFS-3696
> Project: Hadoop HDFS
> Issue Type: Bug
> Affects Versions: 2.0.0-alpha
> Reporter: Kihwal Lee
> Assignee: Tsz Wo (Nicholas), SZE
> Priority: Critical
> Fix For: 0.23.3, 3.0.0, 2.2.0-alpha
>
> Attachments: h3696_20120724.patch
>
>
> When doing "fs -put" to a WebHdfsFileSystem (webhdfs://), the FsShell goes
> OOM if the file size is large. When I tested, 20MB files were fine, but 200MB
> didn't work.
> I also tried reading a large file by issuing "-cat" and piping to a slow sink
> in order to force buffering. The read path didn't have this problem. The
> memory consumption stayed the same regardless of progress.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
