Coming Up: Benchmarks Of GNU Hurd

[phoronix]

Phoronix: Coming Up: Benchmarks Of GNU Hurd

At Phoronix we benchmark Linux, obviously, but of course we also run some tests of Mac OS X, the *BSDs, Solaris and the OpenSolaris-derived distributions, and even Microsoft Windows when doing a hardware/driver comparison to other platforms. It's the platforms that are supported by the Phoronix Test Suite, and with better mobile device support coming, we'll be delivering Phoronix.com benchmarks there too. But there's yet another new platform target with Lillesand: GNU Hurd. Yes, we have benchmarks running now even under GNU Hurd...

http://www.phoronix.com/vr.php?view=OTY3NQ

[vermaden]

MINIX

Do not forget to add MINIX3 to the GNU Hurd benchmarks (http://minix3.org).

[FunkyRider]

I think it would be necessary to build an ancient machine and run the benchmarks on the real hardware instead of under VMs.

[cb88]

Yeah the power management from you host OS can mess up all your numbers ... I saw this when running Haiku which you could also add its a hybrid monolithic kernel unlike Minux and Hurd

Haiku's newer hardware support is advancing pretty fast too... they have a "get it working" mentality unlike HURD guys and they do a good job of gettting it working too.

MICHAEL! WHEN YOU DO I/O INTENSIVE TESTS ON HAIKU... DO IT ON A PARTITION WITH QUERYS DISABLED!

The querys are only useful for media which Haiku rocks at handling and the files don't change much but when you do development you should do it on a drive that had querys disabled for improved filesystem speed as it doesn't have to update the querys synchronously.

All that said HURD is pretty interesting and cool IMO even if development is quite slow I have been watching it for a few years myself

[staalmannen]

Plan9/APE

Plan9 with php compiled under APE would be another interesting target. Since php seems to be pretty portable and is only C (no C++) as far as I have understood, it might be possible to get up there.
I once tried to get it compiled under APE (in a 9vx environment), but I failed (mostly because I am such a noob on Plan9)

For VMs: Anyone tried to run PTS under the Parrot php (Pipp)?

[mitcoes]

XEN would be a good test too and BARRELFISH

I have just read that Xen will use GPU hardware directly, benchmarking it versus the native OS installation results would be a good idea.

And making a BASE, for example Mac Mini as 100 or 1000 to put the results in a relative way would be other great idea.

I would like to see a list from smartphones to Intel/AMD server processors as Toms Hardware does, with this openbenchmark index instead of futuremark or 3dmark.

An a ioquake test with pro settings as this "pro settings" for quake3

http://www.fatal1ty.com/downloads/configs/quake-3/config-cfg/download-358

that is like pro gamers play an where linux has advantages over MS WOS.

And of course others with "full settings" whrw MS WOS has advantage at higher resolutions.

And add 16:9 aspect ratio resolutions now that 4:3 aspect ratio monitors are disapearing of the market. 1280x720 and 1920x1080 - 720p and 1080p - that will be the main resolutions in a near future.

Last but not least, even it is an experimental OS, only for experiments for Phoronix would be a great idea to test also BARRELFISH because of its experimental multiprocessor nature.

[droidhacker]

I think I'll voice my opinion on this one as well... forget the VM, those results are pointless. Find some hardware that it *will* run on, and do *BOTH* Linux *AND* Hurd on that same hardware.

Not that benchmarks matter to begin with (everything is so close that the minor differences are easily ignored), but if you really do insist on benchmarking, at least do it in a way that the numbers will actually make sense.

[cl333r]

Who is planning to use Hurd for serious?

[frantaylor]

Who is planning to use Hurd for serious?

GNU Hurd has a really unique architecture that's going to take a long time for the conventional computer market to understand and grow around. The concepts of "applications" and "drivers" are totally different and it's going to take time for developers to wrap their heads around these new ways of doing things.

As far as the "for serious" part goes, it is ALREADY being used for "serious" thinking about what it means to have a secure operating system, and to most people that is a far more "serious" application than "does it run Firefox"

[locovaca]

I think I'll voice my opinion on this one as well... forget the VM, those results are pointless. Find some hardware that it *will* run on, and do *BOTH* Linux *AND* Hurd on that same hardware.

Not that benchmarks matter to begin with (everything is so close that the minor differences are easily ignored), but if you really do insist on benchmarking, at least do it in a way that the numbers will actually make sense.

If I had to guess, he's just doing testing and fixing of PTS under a VM instead of a physical box.