| To: | Michael Monnerie <michael.monnerie@xxxxxxxxxxxxxxxxxxx> |
|---|---|
| Subject: | Re: LWN.net article: creating 1 billion files -> XFS looses |
| From: | Christoph Hellwig <hch@xxxxxxxxxxxxx> |
| Date: | Thu, 19 Aug 2010 08:05:12 -0400 |
| Cc: | xfs@xxxxxxxxxxx |
| In-reply-to: | <201008191312.49346@xxxxxx> |
| References: | <201008191312.49346@xxxxxx> |
| User-agent: | Mutt/1.5.20 (2009-08-17) |
On Thu, Aug 19, 2010 at 01:12:45PM +0200, Michael Monnerie wrote: > The subject is a bit harsh, but overall the article says: > XFS is slowest on creating and deleting a billion files > XFS fsck needs 30GB RAM to fsck that 100TB filesystem. > > http://lwn.net/SubscriberLink/400629/3fb4bc34d6223b32/ The creation and deletion performance is a known issue, and too a large extent fixes by the new delaylog code. We're not quite as fast as ext4 yet, but it's getting close. The repair result looks a lot like the pre-3.1.0 xfsprogs repair. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: LWN.net article: The end of block barriers, Christoph Hellwig |
|---|---|
| Next by Date: | Re: [PATCH 2/8] xfs: ensure we mark all inodes in a freed cluster XFS_ISTALE, Christoph Hellwig |
| Previous by Thread: | LWN.net article: creating 1 billion files -> XFS looses, Michael Monnerie |
| Next by Thread: | Re: LWN.net article: creating 1 billion files -> XFS looses, Michael Monnerie |
| Indexes: | [Date] [Thread] [Top] [All Lists] |