Hello,

I have been running several benchmarks against 2.6.12-rc1 with differing
levels of CKRM config options on both x86 and ppc64. I finally (due to
the benchmarking system, not due to CKRM) was able to get consistent
results for the 5 key cases for CKRM on both x86 and ppc64:

2.6.12-rc1 stock kernel
2.6.12-rc1 + Gerrit's current stack of patches + CONFIG_CKRM=n
2.6.12-rc1 + Gerrit's current stack of patches + CONFIG_CKRM=y
2.6.12-rc1 + Gerrit's current stack of patches + Chandra's Memory RC
        patches + CONFIG_CKRM=n
2.6.12-rc1 + Gerrit's current stack of patches + Chandra's Memory RC
        patches + CONFIG_CKRM=y

The benchmarks I ran were (followed by the parameters passed in parens):

kernbench -- Workload generated by compiling kernels
        (kernbench $num_cpus 5 -m 2)
tbench -- Simulates a Samba server _without_ filesystem calls
        (tbench both 1)
dbench -- An FS benchmark that generates load patterns similar to those
        of the commercial Netbench benchmark.
        (dbench /tmp 4)
SPECjbb -- Simulation of a Java powered e-commerce website
        (specjbb 1700 4 1 3 19 30 120)
SDET -- Workload created by parallel execution of common UNIX commands
        (sdet "1 4 16 64")
ReAIM (x86 only) -- A multiuser benchmark that tests and measures the
        performance of open system multiuser computers.
        (reaim /tmp/diskdir workfile.short 1 10 2)
Contest (x86 only) -- Designed to test system responsiveness by running
        kernel compilation under a number of different load conditions.
        (no args)

The test machines were:

ppc64: 8-way pSeries (1.5 GHz Power4) with 64 GB RAM
x86: 4-way NUMA-Q (360 MHz PII) with 3GB RAM

I will be sending a total of 12 benchmark-related emails, one per
combination of architecture and benchmark.

The results are in a "raw" format, i.e. exactly what the benchmark
outputs; I figure the discussion of comparisons should occur on the
list.

I realize 2.6.12-rc1 is slightly dated, but it's the best I can do right
now. Hopefully, I can get updated results once I've re-run the tests
against 2.6.12-rc2.

Please ask questions/request more details, if you need them.

Thanks,
Nish


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
ckrm-tech mailing list
https://lists.sourceforge.net/lists/listinfo/ckrm-tech

Reply via email to