Message 202116 - Python tracker

Message202116

Author	vstinner
Recipients	christian.heimes, djc, ezio.melotti, gregory.p.smith, meador.inge, pitrou, python-dev, rhettinger, serhiy.storchaka, vstinner
Date	2013-11-04.10:56:49
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1383562610.53.0.948661058956.issue16286@psf.upfronthosting.co.za>
In-reply-to

Content
Results of benchmarks using compare_hash-3.patch: $ time ../benchmarks/perf.py -r -b default ./pythonorig ./pythonhash INFO:root:Skipping benchmark slowspitfire; not compatible with Python 3.4 INFO:root:Skipping benchmark slowpickle; not compatible with Python 3.4 INFO:root:Skipping benchmark slowunpickle; not compatible with Python 3.4 INFO:root:Skipping benchmark spambayes; not compatible with Python 3.4 Running 2to3... Running django_v2... Report on Linux smithers 3.9.4-200.fc18.x86_64 #1 SMP Fri May 24 20:10:49 UTC 2013 x86_64 x86_64 Total CPU cores: 8 ### 2to3 ### Min: 6.358000 -> 6.055000: 1.05x faster Avg: 6.407600 -> 6.179800: 1.04x faster Significant (t=3.53) Stddev: 0.04311 -> 0.13785: 3.1979x larger ### nbody ### Min: 0.219029 -> 0.212477: 1.03x faster Avg: 0.224940 -> 0.219248: 1.03x faster Significant (t=10.13) Stddev: 0.00373 -> 0.00420: 1.1288x larger The following not significant results are hidden, use -v to show them: django_v2. At least, Python is not slower with the patch :-) I'm surprised that the benchmark shows a difference. nbody benchmark is focused on float numbers. I checked with gdb, nbody benchmark does not call any Unicode comparison function.

Content

Results of benchmarks using compare_hash-3.patch:

$ time ../benchmarks/perf.py -r -b default ./pythonorig ./pythonhash 
INFO:root:Skipping benchmark slowspitfire; not compatible with Python 3.4
INFO:root:Skipping benchmark slowpickle; not compatible with Python 3.4
INFO:root:Skipping benchmark slowunpickle; not compatible with Python 3.4
INFO:root:Skipping benchmark spambayes; not compatible with Python 3.4
Running 2to3...
Running django_v2...

Report on Linux smithers 3.9.4-200.fc18.x86_64 #1 SMP Fri May 24 20:10:49 UTC 2013 x86_64 x86_64
Total CPU cores: 8

### 2to3 ###
Min: 6.358000 -> 6.055000: 1.05x faster
Avg: 6.407600 -> 6.179800: 1.04x faster
Significant (t=3.53)
Stddev: 0.04311 -> 0.13785: 3.1979x larger

### nbody ###
Min: 0.219029 -> 0.212477: 1.03x faster
Avg: 0.224940 -> 0.219248: 1.03x faster
Significant (t=10.13)
Stddev: 0.00373 -> 0.00420: 1.1288x larger

The following not significant results are hidden, use -v to show them:
django_v2.


At least, Python is not slower with the patch :-) I'm surprised that the benchmark shows a difference. nbody benchmark is focused on float numbers. I checked with gdb, nbody benchmark does not call any Unicode comparison function.

History
Date	User	Action	Args
2013-11-04 10:56:50	vstinner	set	recipients: + vstinner, rhettinger, gregory.p.smith, pitrou, christian.heimes, djc, ezio.melotti, meador.inge, python-dev, serhiy.storchaka
2013-11-04 10:56:50	vstinner	set	messageid: <1383562610.53.0.948661058956.issue16286@psf.upfronthosting.co.za>
2013-11-04 10:56:50	vstinner	link	issue16286 messages
2013-11-04 10:56:49	vstinner	create