FreeBSD/PC-BSD 10.2 vs. Ubuntu 15.04/15.10 Benchmarks

[phoronix]

Phoronix: FreeBSD/PC-BSD 10.2 vs. Ubuntu 15.04/15.10 Benchmarks

It's been a while since last running any BSD vs. Linux benchmarks, so I've started some fresh comparisons using the latest releases of various BSDs and Linux distributions. First up, as for what's completed so far, is using the FreeBSD-based PC-BSD 10.2 compared to Ubuntu 15.04 stable and the latest development release of Ubuntu 15.10.

http://www.phoronix.com/vr.php?view=22232

[Shnatsel]

The "Timed HMMer search" benchmark results are very weird. BSD winning over Ubuntu 2x in CPU performance with both compilers hints that something somewhere is very wrong - either in the timing functionality on BSD or in some CPU-related code on Linux.

[peppercats]

Would have preferred Fedora tests after the Ubuntu regressions benchmark the other day. Interesting finds nonetheless, I wonder how much the compilation win is due to file access/read speed of zfs vs ext4.

[monraaf]

After the article yesterday about the success with Radeon KMS under PC-BSD, why were these tests run using the VESA DDX instead of ATI/Radeon DDX?

[cjcox]

Isn't BSD dead? (just trying to be the first to post it)

[Dick Palmer]

The "Timed HMMer search" benchmark results are very weird. BSD winning over Ubuntu 2x in CPU performance with both compilers hints that something somewhere is very wrong - either in the timing functionality on BSD or in some CPU-related code on Linux.

Thought the same. Along with compiling PHP: BSD "winning" by 200%-300% there seems far worse than can be explained by the filesystem difference. There were similarly weird benchmarks a couple of days ago which had me wondering about a kernel (scheduler?chipset driver?) problem. That was a severe slowdown in Ubuntu 15.10 compared with 15.04 on the computational tests. It was on the same 20 core Xeon system as this test. A comment suggests it wasn't reproduced on a 2 core Opteron VM. Something's definitely amiss somewhere.

http://www.phoronix.com/scan.php?page=news_item&px=Ubuntu-15-Benchmarks

[andre30correia]

freebsd and others are good if you don't use it as desktop

[BSDude]

This is why one should use different OSs in production, BSD, GNU/Linux, Solaris.

[pal666]

Along with compiling PHP: BSD "winning" by 200%-300% there seems far worse than can be explained by the filesystem difference.

that one is easy to explain. they have compiler which can produce slow programs fast

[SystemCrasher]

freebsd and others are good if you don't use it as desktop

...and do not use it as server, due to lack of filesystems choice, lack of production-quality VMs & containers and awkward package management. And do not use it in embedded, due to poor hardware support and weird approaches all over place. Supercomputers? Has it got GPU compute support on wheels? Uhm, what else I've forgot?

And on side note, comparing ZFS vs EXT4... well, it can be done, but its like comparing heavy truck vs bike. I guess it makes more sense to do ZFS vs Btrfs test run. It can also be interesting idea to give a shot to ZFS on Linux to see how it performs against these two. And as for ext4, I guess it should be rather tested vs UFS (I can guess the results though).