Solaris 11 Struggles Against Linux Distributions

[phoronix]

Phoronix: Solaris 11 Struggles Against Linux Distributions

For some interesting benchmarks to share before the start of the weekend, here's some recent test results conducted at Phoronix that's comparing Oracle Solaris 11 Express, Ubuntu 12.04 LTS, CentOS 6.2, Debian GNU/kFreeBSD Wheezy, and Fedora 17.

http://www.phoronix.com/vr.php?view=17607

[uid313]

Linux didn't kill Solaris

Solaris struggles against Linux, but it wasn't Linux that killed Solaris - it was Oracle.

Solaris could stand by its own merit. It was backed by a huge technology company (Sun Microsystems), it had the world's most advanced file system, it had great technology such as DTrace, etc.
It was active and had great community participation through OpenSolaris.

Sun was highly regarded in the technical community due to being a huge supporter of open source. Java, OpenSolaris, NetBeans, OpenOffice, VirtualBox and lots of other software. Not to mention open source hardware through the OpenSPARC project.

Then Oracle comes a long, buys Sun and manages to fuck every single thing up.

[LinuxID10T]

Then Oracle comes a long, buys Sun and manages to fuck every single thing up.

+sideways fucking 8

[kraftman]

Solaris struggles against Linux, but it wasn't Linux that killed Solaris - it was Oracle.

No, it was Linux. This is the reason why SUN is dead. Oracle just bought a dead horse, but what's more important they were mainly interested in java not in Solaris.

[jvillain]

Solaris struggles against Linux, but it wasn't Linux that killed Solaris - it was Oracle.

Really? The Solaris guys are still in denial? Go take a look at the stock price graphs. Sun was over long before Oracle came along. Not that I am defending Oracle in any way or form.

The story seems to tie performance to stricktly to the compiler with out taking any thing else into consideration. It would also help if we knew what each of the tests was actually testing.

[ryao]

The story seems to tie performance to stricktly to the compiler with out taking any thing else into consideration. It would also help if we knew what each of the tests was actually testing.

This is true. For instance, the FreeBSD kernel in Debian is compiled with -O1, which hurts performance.

[BlueJayofEvil]

I still wonder from time to time what would've happened if the OpenSolaris code had been released as GPLv3 as some rumors had speculated.
Solaris had some interesting tech in it, but overall it was very hardware-limited. You had to basically buy hardware for it. Linux was already miles ahead of it in that regard.

Any chance Solaris had to shine is long gone. Oracle was just the final nail in the coffin.

[kraftman]

This is true. For instance, the FreeBSD kernel in Debian is compiled with -O1, which hurts performance.

And kfreebsd in FreeBSD is optimized with what flag?

[ryao]

And kfreebsd in FreeBSD is optimized with what flag?

It uses -O2 with GCC 4.2.y. You can compile it with Clang for a performance boost.

[nslay]

The SciMark Monte Carlo results look suspicious. All that should be doing is drawing random numbers and it wouldn't be using OS-features like cryptographic random devices to do so either.

Why is Linux (CentOS) performing substantially worse than Linux (Ubuntu)?

The Solaris results are also unusual in that it performs poorly on tests that have nothing to do with the OS. I'd be really curious about performance if benchmarks were compiled by suncc. However, that wouldn't be a fair comparison.