[
https://issues.apache.org/jira/browse/HDFS-2006?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13997767#comment-13997767
]
Chris Nauroth commented on HDFS-2006:
-------------------------------------
{quote}
I'll also note that many existing FS have arbitrary limits on the size of an
xattr (e.g. page size) and that seems to work fine. I think it's unlikely we'd
need to support large xattrs either, especially since (as Nicholas points out)
we'd need to implement some special storage mechanism to avoid storing large
xattrs with the rest of the FS metadata.
{quote}
This is a good point. In other file systems that do support large xattrs, the
implementation involves allocating extra blocks to store the xattr data outside
the inode struct, and then clients need to read those blocks on access. This
approach would be much less effective in a distributed file system, where
fetching extra blocks incurs network latency.
> ability to support storing extended attributes per file
> -------------------------------------------------------
>
> Key: HDFS-2006
> URL: https://issues.apache.org/jira/browse/HDFS-2006
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Affects Versions: HDFS XAttrs (HDFS-2006)
> Reporter: dhruba borthakur
> Assignee: Yi Liu
> Attachments: ExtendedAttributes.html, HDFS-XAttrs-Design-1.pdf,
> HDFS-XAttrs-Design-2.pdf, HDFS-XAttrs-Design-3.pdf,
> Test-Plan-for-Extended-Attributes-1.pdf, xattrs.1.patch, xattrs.patch
>
>
> It would be nice if HDFS provides a feature to store extended attributes for
> files, similar to the one described here:
> http://en.wikipedia.org/wiki/Extended_file_attributes.
> The challenge is that it has to be done in such a way that a site not using
> this feature does not waste precious memory resources in the namenode.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
