> On 16 Jul 2007 07:03:03 -0700, in bit.listserv.ibm-main you wrote: > > >> Even with PCs, the old standard measurements of processor speed are no > >> longer the selling points that they used to be. There are too many > >> variables that effect general performance. > >> > > > >I think, however, that it is fair to say that peformance is always > >important, even for IBM. The relative power of each new generation of > >mainframe system is part of the sales pitch. > > But sell using meaningful measurements. MIPS isn't a meaningful way > of comparing a mainframe with an alternative. >
Yes, that's why I was suggesting something like SPECjbb. This is an 'industry standard' Java throughput benchmark "SPECjbb2005 is SPEC's benchmark for evaluating the performance of server side Java. Like its predecessor, SPECjbb2000, SPECjbb2005 evaluates the performance of server side Java by emulating a three-tier client/server system (with emphasis on the middle tier). The benchmark exercises the implementations of the JVM (Java Virtual Machine), JIT (Just-In-Time) compiler, garbage collection, threads and some aspects of the operating system. It also measures the performance of CPUs, caches, memory hierarchy and the scalability of shared memory processors (SMPs)." I believe that server side Java is now one of the typical workloads on a mainframe, so this could provide an apples-to-apples comparison if one has the resources and inclination to do so. Regards, Dean ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
