Author nirai
Recipients DazWorrall, aconrad, alex, andrix, brian.curtin, carljm, coderanger, cool-RR, dabeaz, djc, donaldjeo, durin42, eric.araujo, eric.smith, flox, gregory.p.smith, jcea, jhylton, karld, kevinwatters, konryd, larry, loewis, mahmoudimus, movement, neologix, nirai, pitrou, rcohen, rh0dium, tarek, thouis, ysj.ray
Date 2010-05-16.19:16:34
SpamBayes Score 0.00375674
Marked as misclassified No
Message-id <1274037396.45.0.144969205351.issue7946@psf.upfronthosting.co.za>
In-reply-to
Content
A link to ccbench results comparing old GIL, old GIL with long check interval, new GIL and BFS:
http://bugs.python.org/file17370/nir-ccbench-linux.log

Summary:

Results for ccbench latency and bandwidth test run on Ubuntu Karmic 64bit, q9400 2.6GHz, all Python versions built with computed gotos optimization.

Old GIL:
Hi level of context switching and reduced performance. 
~90ms IO latency with pure Python CPU bound background threads and low IO bandwidth results.

Old GIL with sys.setcheckinterval(2500) as done by Zope:
Context switching level back to normal.
IO latency shoots through the roof. ~950ms (avg) is the maximum recordable value in this test since CPU load duration is 2sec.

New GIL:
The expected 5ms wait related IO latency and low IO bandwidth.

BFS patch:
Behaves.
History
Date User Action Args
2010-05-16 19:16:37niraisetrecipients: + nirai, loewis, jhylton, gregory.p.smith, jcea, pitrou, movement, larry, eric.smith, kevinwatters, tarek, djc, karld, carljm, coderanger, durin42, eric.araujo, alex, andrix, konryd, brian.curtin, flox, DazWorrall, cool-RR, rh0dium, rcohen, dabeaz, mahmoudimus, aconrad, ysj.ray, neologix, thouis, donaldjeo
2010-05-16 19:16:36niraisetmessageid: <1274037396.45.0.144969205351.issue7946@psf.upfronthosting.co.za>
2010-05-16 19:16:34nirailinkissue7946 messages
2010-05-16 19:16:34niraicreate