Myrna van Lunteren <[EMAIL PROTECTED]> wrote: > would it make sense & be ok to point to these tests from the 'tests results' > web page?
It would definitively be ok to point to these tests. I leave it to someone else to decide if it "makes sense" :-) ..olav > On 2/17/06, Olav Sandstaa <[EMAIL PROTECTED]> wrote: > > > > Kathey Marsden <[EMAIL PROTECTED]> wrote: > > > Olav Sandstaa wrote: > > > >The results from the performance regression test I run are now > > > >available on the following Web page: > > > > > > > > http://home.online.no/~olmsan/derby/perf/ > > > > > > Thanks Olav. This is so wonderful to have this net and great to see the > > > jump for Network Server. Thanks Knut and Dyre! > > > A couple questions. > > > > > > What is the baseline date? I didn't see the date but am probably just > > > missing it. > > > > It was not included on the Web page, but I did mention it in the email > > I sent. The baseline is based on the repeated runs of the performance > > test on the 10.1.1.0 release. I have updated the Web page with > > information about which version we use as baseline. > > > > > Might it be possible to take snapshots of memory usage and record that > > > information as well to track any glaring memory usage jumps? > > > NetworkServer control runtimeinfo offers the following info: > > > > > > # Connection Threads : 1 > > > # Active Sessions : 1 > > > # Waiting Sessions : 0 > > > > > > Total Memory : 2031616 Free Memory : 1570368 > > > > This sounds like a very good idea. I will include it on my list of > > possible improvements, but unfortunately it seems like I have some > > other higher priority "itches" comming up so it might take some time > > until I get to include it in the tests. > > > > ..olav > >
