Eight-Way BSD & Linux OS Comparison

[RahulSundaram]

For example: http://www.unix-experience.fr/2013/2....yC9IKPC0.dpbs

Can you be more specific? I don't see any security issues there. Only a db specific fs benchmark of 3.2 kernel. That isn't worth much really.

[eLDST0RM]

regarding ext4 being insecure:

For example: http://www.unix-experience.fr/2013/2....yC9IKPC0.dpbs

I don't get it. The link is to some tests that show that performance on a database improves if you turn off barriers on ext4. I don't understand how you can take that fact and use it to conclude that ext4 is inherently unsafe.

[enfocomp]

Very nice, good to see BSD compared with some Linux distros. Thank you very much for these benchmarks, Michael!

Surprised to see that Ubuntu is about the same speed, if not faster, then MINT. I'm glad that MINT was included, because lots of people still have the perception that MINT is substantially faster than Ubuntu, but these results prove otherwise. It seems that Ubuntu and Unity (version 7) are not the pile of crap that some people will lead you to believe!

For the people bitching about the tests used here....why don't YOU do some benchmarking and post the results online if you want to see a specific test or distro? I guess it's easier to bitch rather then doing it yourself though, lol.

[Sergio]

Filesystems take time to mature and using a filesystem used by most distributions as default makes better when testing across operating systems unless you are running a fs benchmark. Also you might want to provide some references on why you consider Ext4 insecure.

For example: http://www.unix-experience.fr/2013/2....yC9IKPC0.dpbs

[RahulSundaram]

Why use crappy and insecure (and hence extremely fast) ext4? Would've been nice if at least one Linux used btrfs...

Filesystems take time to mature and using a filesystem used by most distributions as default makes better when testing across operating systems unless you are running a fs benchmark. Also you might want to provide some references on why you consider Ext4 insecure.

[0xBADCODE]

One of most annoying misfeatures of Phoronix benchmarks is...

The absolutely most annoying misfeature of Phoronix benchmarks to the date is that some results could miss some OSes without explanation given.

Last 5 benchmarks show just 6 competitors instead of 8. So no "8-way". Where all those BSD-based ones? Are they off the track for some reason, or something? And apache benchmark lists just 5 competitors. What happened to another 3? They failed to run Apache? Aw, really? If they are, I think it's worth to mention it, to say the least. Don't you think it's good idea to explain where are all those missing competitors right after appropriate benchmarks? For example I'm curious what BSDs would show in Xonotic on your HD46xx card. Yet there is no results and no any comments why there is no results. Omitting half of benchmark results without explanation is one of biggest misfeatures of those benchmarks to the date.

[timothyja]

I'm not one to criticise, but most would usually base their decision on technical merit, or at the closest to the philosophy point, licensing.
That is unless you're trying to make a personal freedom box that only contains FOSS software and can only access FOSS repositories, but few to no companies actively go out of their way to make such a setup.

You missed one of the philosophy points. The reason I switched to Mageia is that its backed by a Non-profit organisation rather than a company, where decision making is out in the open including financial reports. I find this extremely appealing knowing were my distro is heading and a having a REAL community focus rather than the circus that Ubuntu has become lately.

[intellivision]

You know that some people choose their distribution philosophically, don't you?

I'm not one to criticise, but most would usually base their decision on technical merit, or at the closest to the philosophy point, licensing.
That is unless you're trying to make a personal freedom box that only contains FOSS software and can only access FOSS repositories, but few to no companies actively go out of their way to make such a setup.

[Sergio]

Why use crappy and insecure (and hence extremely fast) ext4? Would've been nice if at least one Linux used btrfs...

[brosis]

Was surprised to see CentOS be fastest on Nexuiz, considering it focus on server and stability.

Khem,.. 2.6.x kernel; 2.1 Mesa; couldn't antialias properly and was just skipping, unlike more modern competition that did their homework. BSD can't game at all. No surprises here.