Re: TCP benchmarks -Reply

Nils I agree. It is a constant source of amazement to me that such a large and established company as Oracle continue to use what basically amounts to lies in their marketing. Sooner or later they must realise that the community has wised up to them, you would think. -- -- Peter Tashkoff NZ Kiwifruit Project Consultant/DBA Marketing Board. Standard Disclaimers apply This posting may not be used by any commercial organisation to vilify another. >>> Nils Myklebust <Nils.Myklebust@ccmail.telemax.no> 13/June/1996 07:17pm >>> David Williams <djw@smooth1.demon.co.uk> wrote: ... snip ... > Does not suprise me - Oracle REALLY ANNOYS ME with this crap. And in the may 20th 1996 issue of ComputerWorld (US) they have done it again: "Best ever TPC-C Performance by database vendor" Oracle 7, Digital AlphaServer 8400 TrueCluster 30,390 Tandem, Himalaya K 10000-112 20.918 Informix, Digital AlphaServer 8400 13,646 First: The Tandem results where withdrawn on April 12th. Maybe it was too late for the deadline of the add? More interesting: The machines used by Informix and Oracle where very different. Oracle used a 4 node clustered machine with 32 processors, while Informix used a single SMP machine with 10 CPUs. Yes, it may be the fastest test, but Informix may have the best results. Their price/performance where $277 versus Oracles $305 per tpmC and a test on the same machine may show Informix even faster. And from the add again: "If crashing at low speed seems OK, call Informix or Sybase. They just haven't been able to make multiple computers work together yet." Anybody who read this group should be able to judge by themselves. The sad thing is the rest of the world may not be initially. Hopefully bad talking competitors will one day fire back. One good thing from this is Oracle has started seeing Informix as a real competitor.