Re: [Gems-users] simulation time

[Abdullah Kayi <apokayi@xxxxxxxxxxxxxx>]

Hi Byn,

I just want to follow up on the topic you brought up. How many processors does your target machine include? Are you using Opal? Although I do not currently use PARSEC, but I am in general have "simulation time" issues with 64-processor targets. Has anyone compared the simulation time difference between 16-core vs 64-core targets for the same benchmark? 

Regards,

Abdullah

On Sat, Nov 27, 2010 at 9:59 AM, Byn Choi <bynchoi1@xxxxxxxxxxxx> wrote:

Hello, everyone-

I'm wondering if I can make a general inquiry on what your experience has been with GEMS in terms of simulation time. It does depend heavily on the application being timed, but my experience has been that for some benchmarks, even with relatively small input, it can take days. For most Parsec apps, it seems that simlarge input is out of reach for GEMS simulation.

I'm wondering if this is "business as usual" or maybe has to do with some weirdness in my setup? Could anyone share his/her story on experiences with GEMS simulation time?

Thanks,

---
Byn Choi
Ph.D. Candidate in Computer Science
University of Illinois, Urbana-Champaign

_______________________________________________
Gems-users mailing list
Gems-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/gems-users
Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/" to your search.