Hello,
I would like to observe different CPU performance metrics in time.
Basically, I would like to treat the metrics as a time-based sequence, just
as when you record sound, for example.
To obtain such metrics for the whole system (i.e., system-wide), I see
basically two approaches (when using pfmon):
a) Print intermediate results
b) Sample a set of events, with one event being the master (timing) event
and the others being the observed events:
Now, my questions are:
* Is the method a) still reasonable, when sampling with fine granularity,
let's say near the scheduler quantum
* Do you think method b) can be better than a) in some circumstances? I
haven't found a "timer tick" event in pfmon, so I would use the count of
instructions. It is then clear that different cores will process different
amount of instructions per-second, so I loose the time semantics, however I
should still have relative timing to the one core.
Thanks
Tom
-------------------------------------------------------------------------
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's still time to save $100.
Use priority code J8TL2D2.
http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
_______________________________________________
perfmon2-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
