FreeBSD 8.0 vs. Ubuntu 9.10 Benchmarks

[phoronix]

Phoronix: FreeBSD 8.0 vs. Ubuntu 9.10 Benchmarks

Canonical will be releasing Ubuntu 9.10 at the end of next month while the final release of FreeBSD 8.0 is also expected within the next few weeks. With these two popular free software operating systems both having major updates coming out at around the same time, we decided it warranted some early benchmarking as we see how the FreeBSD 8.0 and Ubuntu 9.10 performance compares. For looking more at the FreeBSD performance we also have included test results from FreeBSD 7.2, the current stable release. In this article are mostly the server and workstation oriented benchmarks with the testing being carried out on a dual AMD Opteron quad-core workstation.

http://www.phoronix.com/vr.php?view=14219

[RealNC]

FreeBSD got raped :P

[kraftman]

Between FreeBSD 8.0 and Ubuntu 9.10 Alpha 6, the BSD [COLOR=#234865 ! important][COLOR=#234865 ! important]operating [COLOR=#234865 ! important] system[/COLOR][/COLOR][/COLOR] did substantially better. However, this is to no surprise as with Ubuntu 9.10 using the Linux 2.6.31 kernel and EXT4 file-system we have witnessed some performance regressions with SQLite when carrying out database insertions.

It's not sane calling some change a regression. Switch to writeback mode and it should be much faster. Don't spread FUD

http://www.phoronix.com/forums/showp...92&postcount=3

[Ant P.]

Tell that to the average Ubuntu user...

[gunter2]

Between FreeBSD 8.0 and Ubuntu 9.10 Alpha 6, the BSD operating system did substantially better. However, this is to no surprise as with Ubuntu 9.10 using the Linux 2.6.31 kernel and EXT4 file-system we have witnessed some performance regressions with SQLite when carrying out database insertions. UFS on FreeBSD is not impacted by these problems.

This is a different statement than the one here:
http://www.phoronix.com/scan.php?pag...nchmarks&num=4.

In these article you found, that the SQlite regression was just between 2.6.26-2.6.28 and was resolved in 2.6.29. Hence in this article:
http://www.phoronix.com/scan.php?pag...0_alpha1&num=4

you found that the regression was also solved between Ubuntu 9.04 and 9.10.

OK, the system specifications are different between this article and this one (http://www.phoronix.com/scan.php?pag...0_alpha1&num=4) but still the numbers should be roughly on the same level. In that article you measured an SQlite value of 21.11 while in this article you give a value of 771.13. This cannot be explained by the difference in the system specs...

What is going on?

[thefirstm]

It's not sane calling some change a regression. Switch to writeback mode and it should be much faster. Don't spread FUD

http://www.phoronix.com/forums/showp...92&postcount=3

That completely solved the performance regression here (I can't compare it to BSD though, since I don't have it.)

[RealNC]

That doesn't solve the regression. It simply works around it.

[L33F3R]

thx kraftman

also. I honestly expected BSD to have the lead in most of the tests. I was pleasantly surprised.

[Apopas]

Even the slightest degression in a newer version is disgusting.

[kraftman]

That doesn't solve the regression. It simply works around it.

Not sure Maybe such behavior is expected or there are changes in current mode vs previous? I don't see a regression here. Could you point to benchmarks you're according to?

@Gunter2

EDIT:

This is a different statement than the one here:
http://www.phoronix.com/scan.php?pag...nchmarks&num=4.

In these article you found, that the SQlite regression was just between 2.6.26-2.6.28 and was resolved in 2.6.29.

Regression solved in Ext3.

Hence in this article:
http://www.phoronix.com/scan.php?pag...0_alpha1&num=4

you found that the regression was also solved between Ubuntu 9.04 and 9.10.

Ext3 is using different mode (writeback?) in newer Ubuntu and that's why it's faster here. Not a regression this time as Phoronix claims. However, it's little confusing, so maybe I'm wrong?

@Ant. P.

Tell that to the average Ubuntu user...

They don't care

@L33F3R

I never saw *BSD to be faster in real tests. :>

Sorry for editing a lot... my record