<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Alright, trying 10,000 warmup requests, and then 100 benchmark
requests, "production" scenario (gunicorn, one worker, no debug):<br>
<br>
<tt>PyPy3 26.956 [ms]</tt><tt><br>
</tt><tt>CPython 3.4 10.822 [ms]</tt><br>
<br>
So, not much improvement there. Also consider the slight
inconvenience of mandatory manual warmup every deploy:<br>
<br>
* would I need to warm up just the site, every application, or every
template?<br>
* I'd need a prepared session to warm up anything except the login
page<br>
* I'll be running more worker processes, so even more warmup
requests, hopefully they'd get distributed evenly amongst the
workers<br>
<br>
<br>
<div class="moz-cite-prefix">On 07.02.2015 07:56, Omer Katz wrote:<br>
</div>
<blockquote
cite="mid:CAOZCFcCb8_i5u-869QEFtOyKUWL1uQvxu3vjwJc5Yr6=v+1h9Q@mail.gmail.com"
type="cite">
<div dir="rtl">
<div dir="ltr">You need to warm up the JIT first.<br>
</div>
<div dir="ltr">Run the benchmark a 10,000 times on PyPy before
measuring and you'll see the real performance improvement.<br>
</div>
<div dir="ltr">Nevertheless, it does sound like you're hitting a
performance bug(s) somewhere. It's worth investigating.<br>
</div>
</div>
<div class="gmail_extra"><br>
<div class="gmail_quote">
<div dir="ltr">2015-02-07 1:12 GMT+02:00 Tin Tvrtković <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:tinchester@gmail.com" target="_blank">tinchester@gmail.com</a>></span>:</div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">Hello,
PyPy folks!<br>
<br>
While trying to speed up one of my Django sites, I noticed a
new version of PyPy<br>
had just been released. So I grabbed a fresh download of
PyPy 3 (since this is<br>
a Python 3 codebase) and tried taking it out for a spin.<br>
<br>
However, as far as I can see, whatever I try PyPy is
consistently slower than<br>
CPython for this.<br>
<br>
Since this is a proprietary site, I've basically ripped out
all the code except<br>
my settings.py and my requirements; and am benchmarking the
Django admin index.<br>
The results are about the same.<br>
<br>
I've set up a small repo that can be used to reproduce the
environment:<br>
<a moz-do-not-send="true"
href="https://github.com/Tinche/PyPy-Django-Playground"
target="_blank">https://github.com/Tinche/PyPy-Django-Playground</a>.
There's additional info in<br>
the README there.<br>
<br>
These tests have been carried out on Ubuntu Trusty, 64-bit.
CPython 3 is the<br>
system Python, 3.4. PyPy has been downloaded from the
official site and<br>
unzipped.<br>
<br>
So what I basically do is set up an admin session, and use
the Django main admin<br>
page. 200 warmup requests, then 100 benchmarked requests,
look at the mean<br>
request time.<br>
<br>
Some results:<br>
<br>
Django's runserver, DEBUG mode:<br>
<br>
PyPy3 485.389 [ms]<br>
CPython 3.4 105.777 [ms]<br>
<br>
Django's runserver, no debug:<br>
<br>
PyPy3 44.661 [ms]<br>
CPython 3.4 18.697 [ms]<br>
<br>
Gunicorn, 1 worker, no debug:<br>
<br>
PyPy3 28.615 [ms]<br>
CPython 3.4 13.532 [ms]<br>
<br>
I don't exactly claim to be an expert on benchmarking, but
assuming my site<br>
is similar to the Django admin, CPython's gonna be giving me
better performance.<br>
Also the debug runserver performance is kinda worrying.
Nobody's going to be<br>
running this in production, but it makes development a
little slower and more<br>
annoying than it should be.<br>
<br>
Is there anything to make PyPy more competitive in these
kinds of scenarios?<br>
<br>
Kind regards,<br>
Tin<br>
_______________________________________________<br>
pypy-dev mailing list<br>
<a moz-do-not-send="true" href="mailto:pypy-dev@python.org"
target="_blank">pypy-dev@python.org</a><br>
<a moz-do-not-send="true"
href="https://mail.python.org/mailman/listinfo/pypy-dev"
target="_blank">https://mail.python.org/mailman/listinfo/pypy-dev</a><br>
</blockquote>
</div>
<br>
</div>
</blockquote>
<br>
</body>
</html>