Thread: benchmarks
I have someone willing to do a canned benchmark on HP/ux involving Sybase, Oracle and PostgreSQL and publish the results here. Names or no-names. But the benchmarks should be fairly well pre-canned. For example, the postgresql.conf should probably not be out of the box. Who/what can I point him to? I do not follow benchmarks. elein
On Friday 25 July 2003 19:24, elein wrote: > I have someone willing to do a canned > benchmark on HP/ux involving Sybase, > Oracle and PostgreSQL and publish > the results here. Names or no-names. > But the benchmarks should be fairly > well pre-canned. For example, the > postgresql.conf should probably not > be out of the box. Publishing them might be tricky now that you've mentioned the DBs concerned. > Who/what can I point him to? I do not > follow benchmarks. You mean other than: http://www.varlena.com/varlena/GeneralBits/Tidbits/perf.html ;-) Or are you looking for a set of tests? If so perhaps http://osdb.sourceforge.net/ -- Richard Huxton Archonet Ltd
I would suggest doing the same benchmarks that are used by PC Computing or other "pop-trade" magazines that do such comparisons. In fact your tests should mirror as cloesly as possible to tests done by the magazines. Don't worry about having mentioned names. Being transparent is important with benchmarks. Richard Schilling On 2003.07.25 11:24 elein wrote: > I have someone willing to do a canned > benchmark on HP/ux involving Sybase, > Oracle and PostgreSQL and publish > the results here. Names or no-names. > But the benchmarks should be fairly > well pre-canned. For example, the > postgresql.conf should probably not > be out of the box. > > Who/what can I point him to? I do not > follow benchmarks. > > elein > > ---------------------------(end of > broadcast)--------------------------- > TIP 7: don't forget to increase your free space map settings >
On Saturday 26 July 2003 10:32, Richard Huxton wrote: > On Friday 25 July 2003 19:24, elein wrote: > > I have someone willing to do a canned > > benchmark on HP/ux involving Sybase, > > Oracle and PostgreSQL and publish > > the results here. Names or no-names. > > But the benchmarks should be fairly > > well pre-canned. For example, the > > postgresql.conf should probably not > > be out of the box. > > Publishing them might be tricky now that you've mentioned the DBs > concerned. When you publish, make sure to leave out any specfific propriatery db names, use phrases like "The market leading database". > > > Who/what can I point him to? I do not > > follow benchmarks. > > You mean other than: > http://www.varlena.com/varlena/GeneralBits/Tidbits/perf.html ;-) iirc Josh was working with some folks on the -performance list to put together some benchmarking. > > Or are you looking for a set of tests? If so perhaps > http://osdb.sourceforge.net/ or if you want to do some digging, you might see if those eweek "database shootout" benchmarks are still available and appropriate. Robert Treat -- Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL
People, > iirc Josh was working with some folks on the -performance list to put > together some benchmarking. Yes, me, Kevin Brown, Sean Ch. But this is a major project, and I've been too busy with other things to make any progress on it. -- Josh Berkus Aglio Database Solutions San Francisco