Idea: Distributed Benchmarking

An ability to benchmark my computer, upload results to a web-app and see how my computer performs in comparison to other computers.

I am now shopping for a computer and I want to know how much faster it will be than my X61.  I would also like to know that I haven’t set up too many startup programs slowing down performance so that my computer is running significantly slower than other similar configurations out there.  So wouldn’t it be great if I could download a utility, run benchmarking, automatically upload results to a web application, and then see how my computer performs in comparison to other computers out there?  And this would also tell me, if I buy that X201 tablet that I really want, how much faster will it be than the computer I am using now…



Filed under Ideas, Tech

3 responses to “Idea: Distributed Benchmarking

  1. Sounds like something I’d use. For me it’s not so much about buying, though. I use benchmarking more for diagnostics, or, as you suggested when I change system settings etc.

    Because many benchmarks are expensive and/or they are not as flexible and portable as I wanted them, I compiled my own very simple Perl script which basically runs a bunch of integer and floating point calculations and does some memory calls. The big advantage is that it runs on almost any hardware e.g. by means of Linux LiveCDs or on Windows systems by means of Strawberry Perl or ActiveState Perl.

    A simple wrapper (though yet too simple to present here) collects the benchmarks for my own personal comparison.

    • Cool! What does your benchmark measure? I think this could be a good project to approach existing smaller benchmarking companies with. Interested in working on this?

      • Uhm, no :} It’s way too simple. Most everyone interested in benchmarks seems to believe more in full-blown, hardcore, big player commercial suites. And righteously so, I think, if you want to compare e.g. the performance of CPU cache architectures or multi-core benchmarks etc..
        However, personally I found that just in order to have a quick check something like

        for ($i = 1; $i <= ($iterations); $i++) {
        for ($j = 1; $j <= 999; ) {
        $z0=$i * $j++; $z1=$i * $j++;
        $runtime = $endtime-$starttime;

        might not be a proper benchmark but it _does_ help with my daily work.

        Of course, this is single threaded, single core, mostly CPU testing only. For reference, I used (and hopefully improved) someone else's script [ ].

        Concluding, for e.g. HDD benchmarking I do use other tools (the ones which are freely available).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s