Steve Bergman wrote: > On Wed, 2006-09-13 at 13:15 -0500, Steve Bergman wrote: > >> Interestingly enough, I cannot reproduce the writes starving reads >> problem on my desktop Ubuntu Dapper box running "2.6.15-26-k7 #1 SMP >> PREEMPT". >> > > OK. I have been able to test on a freshly installed CentOS 4.4 box > without RAID or LVM and I *do* see a great performance degradation when > bonnie++ is "writing intelligently". (I'm not sure it is as bad as what > I'm seeing on the raid1+lvm2 box, but it is definitely in the ballpark.) > > This box is using a plain old PATA drive. > > So, it's not RAID or LVM2 causing the problem. The fact that the FC5 > box, which is an old celeron 2GHz box with 512MB, does not exhibit the > problem indicates that it is not the CFQ scheduler causing the problem. > > So does 2.6.9 just suck for heavy write IO? Well, RHEL5 will have at least 2.6.17 kernel. That's what was included in public beta 1. If you have any concerns about future RHEL5, now is the right time to download the beta, test it out, and complain. So if the FC5 box works fine, the RHEL5 should work fine too. It would be nice if they decide to switch to 2.6.18 in beta 2 and final for some extra features (like NCQ and hot-plug support for SATA). However, given that 2.6.18 is still in "release candidates" stage, the feature set for RHEL5 more or less getting frozen, and the clock is ticking fast, it is most likely RHEL5 will use 2.6.17. 2.6.18 might be just couple of months (weeks?) too late.