[tahoe-dev] profiling results: what the node does when it is doing nothing

[zooko]

Here's my first results from using the profiler (hotshot) on Tahoe.   
I left the node doing nothing for I thik more than a day.

Happily, the tahoe node doesn't waste a lot of CPU when left idle.

Regards,

Zooko

          2888830 function calls (2887760 primitive calls) in  
141099.963 CPU seconds

    Ordered by: cumulative time
    List reduced from 653 to 16 due to restriction <16>

    ncalls  tottime  percall  cumtime  percall filename:lineno(function)
         1    0.000    0.000 141099.963 141099.963 posixbase.py:219(run)
         1    3.732    3.732 141099.963 141099.963 posixbase.py:223 
(mainLoop)
    145571 141063.776    0.969 141066.878    0.969 selectreactor.py:95 
(doSelect)
    145571    7.753    0.000   25.479    0.000 base.py:630 
(runUntilCurrent)
    141024    1.454    0.000   15.792    0.000 http.py:245 
(_resetLogDateTime)
    141024    9.103    0.000    9.103    0.000 http.py:228 
(datetimeToLogString)
    142689    3.997    0.000    5.283    0.000 base.py:561(callLater)
    145571    1.161    0.000    3.875    0.000 base.py:621(timeout)
    291142    2.285    0.000    3.320    0.000 base.py:612 
(_insertNewDelayedCalls)
    570817    1.887    0.000    1.887    0.000 base.py:161(__le__)
    660652    1.522    0.000    1.522    0.000 <string>:1(setblocking)
      2945    0.047    0.000    1.312    0.000 log.py:38(callWithLogger)
    142689    1.287    0.000    1.287    0.000 base.py:50(__init__)
2962/2954    0.441    0.000    1.262    0.000 log.py:33(callWithContext)
2962/2954    0.017    0.000    0.787    0.000 context.py:58 
(callWithContext)
2962/2954    0.033    0.000    0.768    0.000 context.py:32 
(callWithContext)