On Wed, Jun 15, 2005 at 01:55:58PM +0900, Steven Jenkins wrote:
> In this essay I'm going to attempt, one final time, to demonstrate
> that it is possible to have a useful benchmark. I'm going to do
> so by telling about a real benchmark that we used to solve a real
> problem in the Deep Space Network. But first, some clarifications:
>
Steve, that was a wonderful and fascinating account, thanks very much.

One observation I would make would be that you set up benchmark/test/simulation that was very relevant to your problem domain, you didn't use some industry standard MIPS or TPH or such.  That is the real problem with the type of benchmarks such as spawned the debate here, such things are interesting and I like to look at them, but  that's as far as their usefulness go.


One of the things I do for a living is to migrate municipal databases from various Unix(tm) to Linux on x86 with that proprietary dbms that postgres will hopefully someday kill.  Anyway, the usual transaction per second ratings for various machines are done using a benchmark that uses a standard set of tables from cached memory, so they are completely useless for sizing a cluster of Intel or AMD boxes to replace Unix big iron.  We make our own tests with the client's dbms and software. So we could call that a "benchmark", though it's more of a benchmark/simulation/test type of thing. 


Ralph PJPizza