File-System Benchmarks With The Linux 2.6.34 Kernel

[phoronix]

Phoronix: File-System Benchmarks With The Linux 2.6.34 Kernel

File-system benchmarks have become quite common to Phoronix in the age of EXT4 and Btrfs with these new file-systems driving much of the interest and as we have also been finding the Linux file-system performance to change between kernel releases (and in some cases, the performance has changed a great deal). Most recently we delivered benchmarks of EXT4 vs. Btrfs vs. Reiser4, but now a month later we are back with more Linux file-system benchmarks as we look to see if the disk performance has changed with the Linux 2.6.34 kernel.

http://www.phoronix.com/vr.php?view=14778

[d2kx]

btrfs looks like a winner, even though it did not win most benchmarks. I am looking forward to try it with Ubuntu 10.10, with Kernel 2.6.36 or something.

[kiwi_kid_aka_bod]

Again, this testing was atop a solid-state drive with the default mount settings -- as other tests have shown, using the Btrfs SSD mode can lead to better performance as well as enabling the zlib compression mode.

I quickly scanned through the graphs on the linked page. In just ONE single test did SSD mode improve performance. Indeed it often seemed to cause quite a significant penalty. Surely this makes the above statement completely incorrect!

This is of course not even touching on the fact that the linked article is nearly a year old...

[Michael]

I quickly scanned through the graphs on the linked page. In just ONE single test did SSD mode improve performance. Indeed it often seemed to cause quite a significant penalty. Surely this makes the above statement completely incorrect!

This is of course not even touching on the fact that the linked article is nearly a year old...

Right, more recently the SSD mode provides better gains.

[blagishnessosity]

I wouldn't mind seeing this test repeated with btrfs vs btrfs SSD mode vs btrfs w/zlib vs btrfs w/NOOP or DEADLINE scheduler, so we can get an idea of how big a difference each feature makes.

[mendieta]

I guess it's about time to migrate my /home from xfs to ext4 (/home was already ext4), perhaps leave /data as xfs ..

[kimble85]

I wouldn't mind seeing this test repeated with btrfs vs btrfs SSD mode vs btrfs w/zlib vs btrfs w/NOOP or DEADLINE scheduler, so we can get an idea of how big a difference each feature makes.

That would be very interesting!

[mendieta]

I guess it's about time to migrate my /home from xfs to ext4 (/home was already ext4), perhaps leave /data as xfs ..

I meant to say "(/ is already ext4)"

[Panix]

I think I read if you have kernel 2.6.34, there will be the required packages to ensure that TRIM is working and that partitions will be aligned.

Can anyone confirm this or list what is needed to have a trouble-free experience with a SSD in Linux? Most of the related sites I've found are a bit outdated or seem to indicate there are a lot of configuring/tweaking and I'd rather just plug and play if things are matured and ready to go (with a more recent kernel/latest packages).

[Delgarde]

BTRFS anomalies

Interesting - for the most part, BTFS performs well, if not exceptionally so. But there are just a couple of tests - the database ones - where it lags hugely behind everything else. What makes it so much worse than the others at that particular usage pattern?