Re: TPC-A and -B Benchmarks

In a recent append, danr_at_utoday.com (Dan Richman) writes...
>Does anyone really rely on database benchmarks anyway? If so, how, and to
>what extent? Should the TPC scrap TPC-A and -B? If they're tainted now,
>can they be revived? If so, how? Thanks for any thoughts. This is for a

I would be interested in hearing this story expanded to cover the emerging TPC-C benchmark as well. Many people in the industry have decided that the TPC-A and TPC-B benchmarks do not adequately represent a typical OLTP workload and believe that TPC-C is at least a better approximation. I have heard this opinion before Oracle ever introduced its "discrete transaction".

Only a few companies have published TPC-C results (Informix and IBM at least) and Oracle's lack of published TPC-C results is interesting.

Given the uproar over the "discrete transaction" and the Standish report, I would be interested in hearing the official Oracle viewpoint on the TPC-C benchmark.

Pat Gainer
gainer_at_almaden.ibm.com Received on Sat Jun 26 1993 - 18:09:17 CEST