Re: [Gems-users] Regarding L2 miss rate

[sparsh mittal ISU <sparsh@xxxxxxxxxxx>]

Hello Byn

Thanks for the prompt reply.In terms of the stats reported by ruby, would you explain the term,
"{# of mem accesses actually seen by L2}".

My understanding is that, global l2 miss rate is what I found, namely, L2_misses/(L1Imiss+L1Dmiss).

Thanks

Sparsh

On Tue, Mar 16, 2010 at 11:30 AM, Byn Choi <bynchoi1@xxxxxxxxxxxx> wrote:

Global L2 miss rate : {# of L2 misses}/{# of all mem accesses coming out of cores}
Local L2 miss rate : {# of L2 misses}/{# of mem accesses actually seen by L2}

Reference: Computer Architecture: a Quantitative Approach by John Hennessy and David Patterson

Byn

On Mar 16, 2010, at 11:22 AM, sparsh mittal ISU wrote:

Hello List
I searched the previous list postings a lot, however could not definitely find out the answer. I wanted to know about L2 miss rate. Is it simply L2_misses/(L1Imiss+L1Dmiss) or something else?

--
Thanks and Regards
Sparsh Mittal

_______________________________________________
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.

-- 
Thanks and Regards
Sparsh Mittal
Graduate Student
Electrical and Computer Engineering
Iowa State University, Iowa, USA