On Wed, Jul 06, 2011 at 06:51:07PM +0200, Adi Kriegisch wrote: > * AoE (and iSCSI too) uses a block size of 4K (a size that perfectly fits > into a jumbo frame). So all I/O is aligned around this size. When using a > filesystem like ext4 or xfs one can influence the block sizes by creating > the file system properly.
No, AoE has no block size. It will cram as many sectors as it can into a packet; e.g. if the MTU is 9000, then 17 sectors fit inside it, which does not play well with any kind of alignment. [...] > >From my point of view there are several ways to find the root cause of the > issues: > * try a different RAID level (like 10 or so) > * (re)-try to export the disks to Linux as JBODs. > * try different filesystem and lvm parameters (actually you better write a > script for that... ;-) And if you insist on using parity RAID (i.e. RAID5 or RAID6), then make sure the number of data disks is a power of two. That makes computing various alignments much easier. Gabor ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2d-c2 _______________________________________________ Aoetools-discuss mailing list Aoetools-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/aoetools-discuss
