Am Samstag, den 31.12.2005, 01:39 +0000 schrieb Chris Lightfoot: > On Sat, Dec 31, 2005 at 12:36:50AM +0100, Juraj Holtak wrote: > > I still do not understand, why the guests became that much slow. Could > > somebody explain me why? > > Very probably you're suffering from seeing the real seek > performance of the disks. Switching on the write cache > hides this, at the cost of losing data in the event of a > crash; switching it off make it less likely you'll lose > data in a crash, but significantly shortens the life of > the disks.
Yes. Thats for sure one part of the puzzle. The hardware got slower, but not THAT much slower. The host been 50x to 100x faster and not affected by the uml-guest writes at all (yes. they share the same disks). It had little to no effect at all, to the host write performance, when 2 guest were in 100% io wait and the disks were constantly writing with ~ 2000kbit/s. The host was responsible and I copied over gigs of files just to be sure, to fill up all possible caches. I also don`t think, that the guest filesystems are that much fragmented. They are about 6GB big und filled up to 10%. The mystery isn`t solved yet. juraj ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ User-mode-linux-user mailing list User-mode-linux-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
