Benchmarks of Debian GNU/kFreeBSD vs. GNU/Linux

An anonymous reader writes "The Debian Squeeze release is going to be accompanied by a first-rate kFreeBSD port and now early benchmarks of this port have started coming out using daily install images. The Debian GNU/kFreeBSD project is marrying the FreeBSD kernel with a GNU userland and glibc while making most of the Debian repository packages available for kfreebsd-i386 and kfreebsd-amd64. The first Debian GNU/kFreeBSD benchmarks compare the performance of it to Debian GNU/Linux with the 2.6.30 kernel while the rest of the packages are the same. Results are shown for both i386 and x86_64 flavors. Debian GNU/kFreeBSD may be running well, but it has a lot of catching up to do in terms of speed against Linux."

Comments

Re:Please educate me a bit.

[idiotnot]

Not only that, they're cleanly managed with apt and dpkg. Ports has grown so unwieldy that when I do use FreeBSD these days (admittedly, it's pretty rare), I don't build anything from ports, and use pkgsrc. On the rare occasion that pkgsrc doesn't have something I need, I'll just build it from source manually. But, but, but, but portsupgrade!!!1! It sucks compared to apt/dpkg and pkgsrc. 8, 10 years ago, maybe it was great. But notsomuch anymore.

I also use pkgsrc on Slackware, OpenBSD, and OS X.

Re:Mod Article Flamebait

[ThePhilips]

... you'll almost exclusively be testing usermode code unrelated to the OS.

Nobody cares about what OS itself does.

All people want from an OS is that their application run and run best.

... but you'll mostly be testing 3rd party code and the quality of your compiler's optimizations, not the OS kernel.

The statement is valid for e.g. MS-DOS. But it never was valid for OSs supporting virtual memory and I/O abstraction: the way kernel does things impact application performance quite noticeably. I/O optimization (read-ahead, delayed write back) and virtual memory management (application memory allocation, stack, context switching) all have direct influence on user space performance. And that is what important and is subject of the Phoronix tests.

P.S. My personal gripe with the tests is that they do not attempt to measure latencies. For some I/O-bound applications and also video playback latencies are more important compared to throughput or CPU usage.