Benchmarks Of Debian Etch, Lenny & Squeeze

[phoronix]

Phoronix: Benchmarks Of Debian Etch, Lenny & Squeeze

With Debian 6.0 "Squeeze" set to be released in the coming months, we have decided to run a set of benchmarks looking at the performance of Debian 6.0 across different sub-systems relative to the performance of Debian 5.0 "Lenny" and Debian 4.0 "Etch" to see how this new release may stack up.

http://www.phoronix.com/vr.php?view=15459

[NomadDemon]

mega regress:<

[loonyphoenix]

mega regress:<

Possibly the cost of greater reliability.

[curaga]

Typo on the first page, ext4 should be ext3.

[devius]

Possibly the cost of greater reliability.

I wish someone with money to spare on hard drives would prove that.

[yogi_berra]

mega regress:<

glibc -> eglibc

[loonyphoenix]

glibc -> eglibc

The benchmarks that regressed all had something to do with disk IO. The CPU-oriented tasks increased their performance. So I don't think the eglibc transition has anything to do with it; so I think more careful use of barriers and such to improve data security is responsible for the slow-downs. Or at least I hope so.

[etnlWings]

OT this is the first time I've looked at Phoronix's benchmarks in some time. I'm not sure if they've only just been introduced but I'm loving the inclusion of error margins.

[devius]

OT this is the first time I've looked at Phoronix's benchmarks in some time. I'm not sure if they've only just been introduced but I'm loving the inclusion of error margins.

Where have you been then? There have been lots of articles regarding the new improvements that are coming with PTS 3.0 and all that. That error margin thing is one of those improvements.

[etnlWings]

I follow Phoronix mostly just to keep in-touch with general *nix and X/Mesa developments. I generally avoided all the benchmarking stuff because, well, without some expression of the confidence of the results, the results were next to meaningless; there was no way to determine the significance of a given value. Hell, for all I knew, the values plotted were the result of a single test - or possible outlier. I understandably fell into the habit of just skipping over any feed headlines including the words, "test", or, "benchmark", like I do for any including the word, "Apple".