Hi, Nicolas Boullis wrote: > On a reasonably new host, using as UBD devices logical volumes built > on top of a FC SAN, building a 10GB ext3 filesystem took something > like 50 seconds before my patch, and something like 3 seconds after > it. (I can check the exact values if you wish.)
Hmmm... my memory was making thing much better than they really are. It's around 38 seconds with 2.6.18 (before the "batch I/O requests" change); it's several minutes with unpatched 2.6.24, and it's around 7 seconds with patched 2.6.24... To be more honnest, my patch contains a part similar to Steve VanDeBogart's "ubd does multiple io's when one will suffice" patch. Using 2.6.24 with his patch but still serialized I/O, the time is around 23 seconds. Hence, my patch (with 16 threads per device) make things around 3 times faster. Cheers, Nicolas Boullis École Centrale Paris ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
