[Gc] Performance evaluation

Manuel.Serrano at sophia.inria.fr Manuel.Serrano at sophia.inria.fr
Tue Aug 10 11:34:05 PDT 2010


Hi Hans,

> I'm a little confused about the units and axis labels here.  What
> exactly is being measured in each case?  The 1.02x figures and the like
> are throughput?  The other numbers are time?  bigloo3.2b uses GC7.1 by
> default, bigloo3.4b uses 7.2alpha4 unless stated otherwise?  What does
> "3.2bMHz" mean?
You are absolutely right. This is highly confusing. I have re-used a previous
script to generate the barchart and not correctly changed the labels.

What is measured here is the SYS+CPU time on the same machine (a Linux
core I7, 3.2ghz with 6GB of RAM). The base time is bigloo3.2b + gc7.1.

> On benchmarks like Traverse, the time increased significantly from 7.1
> to 7.2alpha4?  If I'm reading that correctly, is it easy to generate a
> GC log for the two cases?  It's possible that we broke something in
> 7.2alpha4.  But I think we also may have fixed some bugs that could have
> caused 7.1 to allocate way too much heap space, and hence run faster.
Yes. Traverse is the most annoying one but in general it looks like as
if 7.2 was constantly a little slower. Do you want me to investigate on
that particular benchmark. I can study the number of allocations, the overall
allocated memory, the number of GCs.

> The configuration is the same in both cases, particularly with respect
> to thread-support (incl. thread-local allocation, parallel marking) and
> debugging options?
Yes. Exact same thing. 

-- 
Manuel


More information about the Gc mailing list