Re: could use some assistance in tracking down some current issues

On 21-8-2013 11:10, Éric Piel wrote:
> On 20/08/13 20:49, Irmen de Jong wrote:
> :
>>> =========================================
>>> eventloop slow with multiple calls: I don't know how to test.
>>> $ python ./eventloop/client.py
>>> 5*11=55
>>> 'x'*10=xxxxxxxxxx
>>> press enter to do two more calls:
>>> 2*4=8
>>> '@'*3=@@@
>>>
>>> (That looked like less than a second, both multiplex and threaded)
>>
>> I've very recently updated this example to use a loop in the client, that should finish
>> almost instantaneous because it is only 20 iterations. It does that on windows and OSX,
>> but on Linux, it takes much longer with only a few calls/sec

> Yes, I got the new version with a pull this morning. That's fresh! And
> indeed, it's painfully slow on my system (both threaded and multiplexed
> server):

I've just fixed this particular issue. I've been changing Pyro's wire protocol and the
smaller message header and the way I wanted to write the message bytes to the socket,
triggered Nagle's algorithm. This caused the delays. Funny that it only kicked in on
Linux systems and not on OS X or Windows...

Irmen


------------------------------------------------------------------------------
Introducing Performance Central, a new site from SourceForge and 
AppDynamics. Performance Central is your source for news, insights, 
analysis and resources for efficient Application Performance Management. 
Visit us today!
http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk