Re: Fastcode MM B&V 0.41

Hi Robert,

> I don't understand the 94.9% in the second run. How can opening a web page
> affect the memory usage of the B&V tool? Which benchmarks were influenced?

It appears that the 2GB virtual address space that each process gets is not
completely independent from what other things are happening in the system. I
have not really done a complete analysis, but maybe Windows reserves some
user address space (for whatever purpose) when you alt-tab to another
application. Perhaps there are reserved chunks in the 2GB user address space
that is actually shared among all processes.

> Your results seem to suggest that random influences play a much larger
> role than the execution order.

Not really. If you leave the system alone, the memory usage scores stays
very close between benchmark runs (provided you clear the results grid
before each run). Opening up other programs during a benchmark is not
recommended since it obviously affects the speed score and (as the tests
show) also the memory score. The execution order affects the score even if
you do not touch the keyboard.

Regards,
Pierre


.

Relevant Pages

  • Re: sparse polynomial arithmetic
    ... Fateman's benchmark, double-precision coefficients (16-bit ... memory usage, then probably the hash table gets resized but the memory ...
    (sci.math.symbolic)
  • Re: [Full-Disclosure] Firefox 0.92 DoS via TinyBMP
    ... The first time I opened the page memory usage could be seen on Task ... Are you certain this is a vuln in Firefox? ... I get almost identical results opening the same page in IE ...
    (Full-Disclosure)
  • Re: jruby vs mri memory management
    ... Linux 32bit when we benchmarked the Ruby Quiz #157 submissions, ... was not impressed (I didn't check the memory usage though): ... you just had to launch the benchmark to get it. ... JVM). ...
    (comp.lang.ruby)
  • Re: Fastcode MM B&V 0.41
    ... > As the results grid grows, the memory usage scores of the MMs ... If you run a benchmark batch with MMs in the order BucketMM first ... ...
    (borland.public.delphi.language.basm)
  • Excessive Memory Usage with libxml
    ... I've got some basic functionality going, it's not finished, and when ... When I run the benchmark I get weird things happening. ... 10000 iterations were fine ... The memory usage though was stupid at 235Mb Real and 233Mb Virtual ...
    (comp.lang.ruby)