Re: [Gems-users] How to compare performance of two different configurations?

[Dan Gibson <degibson@xxxxxxxx>]

Ruby_cycles measures simulated time. The value of Ruby_cycles is the 
count of the number of times the Ruby event queue is invoked in the 
course of simulation. The Ruby event queue is invoked every 
SIMICS_RUBY_MULTIPLIER cycles (eg. every 2 cycles, see 
ruby/config/rubyconfig.defaults) from the Simics event queue associated 
with Simics processor zero. Each "Ruby cycle" is one simulated cycle of 
the memory system--analogous to one cycle of a logical "memory" clock. 
Ruby_cycles are *not* determined by the number of instructions executed 
on any processor.
With the example you gave, it would seem that your workload took about 
twice as long to execute on two processors, compared to single-processor 
execution.
Regards,
Dan

Liping Xue wrote:
> Hi, I have a quick question about ruby_cycles.
> The Protocol I used is MOESI_CMP_TOKEN;
> I want to compare the ruby_cycles of two configurations:
> one is single processor :
> -----------------------------------------------------
>       ruby0.setparam g_NUM_PROCESSORS 1
>       ruby0.setparam g_PROCS_PER_CHIP 1
> ------------------------------------------------------
> the other is two processors(2 processors on a chip)
> -------------------------------------------------------
>      ruby0.setparam g_NUM_PROCESSORS 2
>      ruby0.setparam g_PROCS_PER_CHIP 1
> -------------------------------------------------------
> And other configuartions are same.
> For the same application, for the first single processor configuration,
> I got "Ruby_cycles: 5074917", for the second 2 processors configuration, I
> got
> "Ruby_cycles: 11123235". How can I compare the performance of the given
> application running on different configurations? I also want to make sure
> how the "Ruby_cycles" are calculated? Does it add up the total cycles of
> two processors? If it is true, can I say that for the given application,
> the performance are alomost same for these two configurations. (Since
> 11123235 /2 is close to 5074917.
>
> Thanks for your help!
>
> -- Liping
>
>
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>