Hide Forgot
+++ This bug was initially created as a clone of Bug #634649 +++ Description of problem: According to multiple sources (http://media.netapp.com/documents/tr-3747.pdf for example) alignment can have huge impact on performance. Our qcow2 is indeed aligned to 64K boundaries, but then XP goes and install itself on sector 63, which means 31.5K into the first qcow2 cluster. That's not that great. See the above NetApp PDF, section 4.4.1 and on, how this can easily be fixed pre-installation. libguestfs should do it via its NFS utilities. This can be later utilized by RHEVM. --- Additional comment from rjones on 2010-09-17 06:03:33 EDT --- Mike Snitzer wrote these documents on the issue: http://people.redhat.com/msnitzer/docs/io-limits.txt http://people.redhat.com/msnitzer/snitzer_rhsummit_2009.pdf
If we do the rebase then we will get the two tools virt-alignment-scan and (updated) virt-resize in RHEL 6.3.
Setting to modified on the basis that libguestfs 1.16.2 is available and contains the new tools (virt-alignment-scan and updated virt-resize).
Technical note added. If any revisions are required, please edit the "Technical Notes" field accordingly. All revisions will be proofread by the Engineering Content Services team. New Contents: Cause: Guest has partitions which are not aligned for best performance. Until around 2008, most guests installed themselves non-optimally like this, and this applies even when installing a pre-2008 guest today. Consequence: The guest suffers a performance penalty, as much as 20%. Fix: libguestfs now provides two tools which can diagnose and fix these guests: virt-alignment-scan and (the updated) virt-resize. Result: Customers can use virt-alignment-scan to identify which guests may suffer from poor performance because of unaligned partitions. Furthermore they can fix these guests by using virt-resize. Customers should read the respective manual pages carefully first.
set to verified, according to the above result, and i didn't see any side-effect
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHSA-2012-0774.html