Comments on: …and statistics

By: ML

ML — Tue, 27 Jul 2010 11:18:42 +0000

While this post is somewhat old, I wanted to add another 2 cents. In our system we have had several problems with malloc performance on Mac OS X. Since we are delivering a library, we don’t want to fiddle too much with the system malloc or the environment. We discovered the problem when we added multi-threading to our library. On Windows and Linux we got a nice speed-up (almost linear), but for Mac OS X the speed-up was much much worse. The reason turned out to be inefficient malloc. Adding a small caching layer for allocated blocks in our library solved the problem, and gave a nice speed-up even for single-threaded programs. On Windows and Linux the layer gave a small but not significant speed-up.

Another problem was for some users that did lots of benchmarking in the same process. The program allocates a small amount of memory each time around, but in their case this was slightly larger each time. This eventually led to an Out of Memory error, even though all memory was released after each round in the benchmark!

By: koyali

koyali — Tue, 03 Mar 2009 09:22:36 +0000

hi this is koyali and most of the they want to know the details about jobs now a days this is common to every one so i will provide some information here please visit.
============
koyali
temping work in london

By: antiquariato

antiquariato — Sat, 02 Dec 2006 21:47:45 +0000

The best way is to http://antiquariato.litect.org/antiquariato.html antiquariato and read all the information available on different antiquariato websites. Go get it!

By: Louis Duran

Louis Duran — Tue, 19 Sep 2006 20:57:37 +0000

Wow, even poli-sci professors at Berkeley know about computing. When’s the last time a poli-sci professor you know compared malloc to gmalloc?

By: Linux Unix » Debunked: Sehkon’s Mac OS X / Ubuntu / WinXP Core Duo benchmarking results

Fri, 01 Sep 2006 12:05:04 +0000

[...] with the same one used in Sehkon’s Windows code yields equal performance on the Mac.read more | digg story Comments are closed. [...]

By: buy flonase

buy flonase — Sun, 16 Jul 2006 10:59:36 +0000

buy flonase

buy flonase
buy flonase – buy flonase
You learn to write as if to someone else because NEXT YEAR YOU WILL BE
“SOMEONE ELSE.”
You are wise, witty, and wonderful, but you spend too much time reading
this sort of trash.

By: Memoria de Acceso Aleatorio » Â¿SerÃ¡ Leopard mÃ¡s rÃ¡pido que Tiger en mi mÃ¡quina?

Thu, 22 Jun 2006 15:37:32 +0000

[...] Opteron Benchmarks Shekhon: Linux versus Mac OS X on Intel Dual Core Ridiculous Fish: â€¦and Statistics Este artí [...]

By: Max Williams

Max Williams — Thu, 25 May 2006 17:35:28 +0000

Nice article. I sorta understood it.

By: penk - Keep on rockin’ in the free world » Blog Archive » ï»¿æœ¬æ—¥æ›¸ç±¤

Thu, 25 May 2006 15:08:02 +0000

[...] ardo.com/archives/7-more-reasons-why-web-apps-fail/ Linux vs. MacOS X Benchmarks Debunked http://ridiculousfish.com/blog/archives/2006/05/16/36/ GNOME 2.15.2 Released! http://mail [...]

By: JSS

JSS — Thu, 25 May 2006 04:38:11 +0000

I have updated my benchmarks. I have significantly improved the efficiency of my code on all platforms, and I now link against dmalloc on OS X. But an OS X performance issue remains.

http://sekhon.berkeley.edu/macosx/

Suggestions are most welcome.

JS.