[Swift-devel] a weekend playing with graphs

[Ioan Raicu]

Hi,
They are basically some simple ploticus graphs based on the Falkon logs, 
and then a web server (and some scripts) that showed the graphs and 
refreshed them every so often.  You'll find things in the falkon SVN two 
directories, falkon/ploticus and falkon/webserver where all the scripts 
and various pieces of code are.   Keep in mind that I ended up making 
separate logs in Falkon that are easy to parse and friendly to ploticus, 
as I found my general human readable logs to be too hard to parse 
reliably. 

Just as a reminder, here are the 3 graphs that get generated 
automatically from the Falkon logs:

   1. http://people.cs.uchicago.edu/~iraicu/research/docs/MolDyn/244-mol-success-8-10-07/summary_graph-med.jpg
   2. http://people.cs.uchicago.edu/~iraicu/research/docs/MolDyn/244-mol-success-8-10-07/executor_graph-high.jpg
   3. http://people.cs.uchicago.edu/~iraicu/research/docs/MolDyn/244-mol-success-8-10-07/task_graph-med.jpg

Here are some sample log files from Falkon (they don't correspond to the 
graphs above):

   1. http://people.cs.uchicago.edu/~iraicu/research/docs/MolDyn/244-mol-failed/Falkon_logs/GenericPortalWS_perf_per_sec.txt
   2. http://people.cs.uchicago.edu/~iraicu/research/docs/MolDyn/244-mol-failed/Falkon_logs/GenericPortalWS_taskPerf.txt

Basically, graph #1 uses log #1, and graphs #2 and #3 uses log #2.

Ioan 

itf at mcs.anl.gov wrote:
> Cool. I encourage you to look at Ioan's tools--I seem to recall them
> showing number of executing tasks, number of executing tasks, number of
> blocked tasks, and maybe some other things--with different colors for
> different sorts of tasks. Seeing as we have those tools, perhaps worth
> using.
>
>
>   
>> http://www.ci.uchicago.edu/~benc/bm1/
>>
>> here's a bunch of graphs that I have got being generated automatically
>> from log files and kickstart records.
>>
>> some of these graphs need kickstart to be enabled (the kickstart ones).
>> some of them need as-yet-uncommitted log changes (the execute, but not
>> execute2 ones).
>>
>> i've also started on some text-based statistics.
>>
>> ideas for useful measurements welcome (both for debugging and for showing
>> (lack of) performance)
>>
>> --
>> _______________________________________________
>> Swift-devel mailing list
>> Swift-devel at ci.uchicago.edu
>> http://mail.ci.uchicago.edu/mailman/listinfo/swift-devel
>>
>>
>>     
>
>
> _______________________________________________
> Swift-devel mailing list
> Swift-devel at ci.uchicago.edu
> http://mail.ci.uchicago.edu/mailman/listinfo/swift-devel
>
>   

-- 
============================================
Ioan Raicu
Ph.D. Student
============================================
Distributed Systems Laboratory
Computer Science Department
University of Chicago
1100 E. 58th Street, Ryerson Hall
Chicago, IL 60637
============================================
Email: iraicu at cs.uchicago.edu
Web:   http://www.cs.uchicago.edu/~iraicu
       http://dsl.cs.uchicago.edu/
============================================
============================================

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/swift-devel/attachments/20070923/5dc3ea9c/attachment.html>