On Dec 08, 2005 09:20 +0000, Holger Kiehl wrote: > Do you have any performance numbers that show how these cards increase > performance? I have tried doing it with a ramdisk and the performance > was not much better, using ext3 and journal on ramdisk. ext2 is still > much faster, here some quick tests with dd: > > 2.6.15-rc1-git2 > > time dd if=/dev/full of=/home/aaaa bs=4M count=4883 > SW Raid 1+0 > ext2 > real 1m11.856s 1m16.050s 1m18.449s > user 0m0.020s 0m0.000s 0m0.010s > sys 0m44.630s 0m44.820s 0m45.870s > > ext3 writeback > real 2m1.511s 1m40.958s 1m43.889s > user 0m0.070s 0m0.000s 0m0.000s > sys 1m33.390s 1m21.470s 1m23.430s > > ext3 writeback+external journal in ram disk > real 2m5.357s 1m41.349s 1m36.228s > user 0m0.010s 0m0.020s 0m0.010s > sys 1m37.500s 1m24.110s 1m21.960s > > Wonder how one can increase the write performance but still having a > journal. You should test a metadata workload that uses the journal a lot (e.g. bonnie file create/unlink) to see if there is a more noticable effect for the external journal. Also, increasing the journal size can have a noticable effect on performance, which is why Ted increased the default journal size for large filesystems to 128MB. To get ext3 performance on-par with ext2 it is possible to use the delalloc+mballoc+extent patches, which Alex+IBM were testing for 2.6.12 a while back. I believe Mingming also worked on delalloc patches for regular block-allocated ext3 filesystems. If you search for "mballoc delalloc" in ext2-devel you will find some benchmark data on this. Hopefully Alex will have some time in the new year to make the new version of the mballoc patch and updates for a newer kernel available for testing. Cheers, Andreas -- Andreas Dilger Principal Software Engineer Cluster File Systems, Inc. - To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
