FW: PDQ

I have 5 CPU VPs, so this shouldn't be the problem. I've been meaning to read a bit more about PDQ - I've just been a bit busy. I touched a little on PDQ in 7.30 I'm now working with 9.21FC1 and have to just touch up on PDQ again. I did set the PDQ parameters in the onconfig-file with the idea of using PDQ for myself, but no environment variables were set, and I actually caught some users with multiple threads, same sid (onstat -u) - very strange to me. I think RTFM is a good idea in my case. It still seems strange though that in one session I actually set PDQPRIORITY and is doesn't use PDQ, and in another I didn't set anything, and it used PDQ (onstat -g mgm). Regards Dirk -----Original Message----- From: Art S. Kagel [mailto:kagel@erols.com] Sent: Wednesday, February 07, 2001 10:20 AM To: Dirk Moolman Subject: Re: PDQ How many CPU VPs do you have? PDQPRIORITY 50 implies that the query in session 27 requires 50% of each resource, memory and CPU VPS, if it cannot get that it will wait until it can. If you only have one CPU VP then session 27 has to wait until it can run all alone in order to get 50% of the CPU VPs since a thread in a VP in non-interruptable. Art S. Kagel Dirk Moolman wrote: > > This is a multi-part message in MIME format. > > ------=_NextPart_000_005A_01C08E0D.B35BA2A0 > Content-Type: text/plain; > charset="iso-8859-1" > Content-Transfer-Encoding: 7bit > > Hi, I just ran into something very strange. > > I am running to queries on my system (doing my weekly maintenance .....) > > For sid 27, I set PDQPRIORITY=50, PSORT_NPROCS=5 > For sid 87 I didn't set anything > > When I run "onstat -g mgm", sid 87 shows as the active query and not sid > 27. Why is this, or don't I understand PDQ properly ? > > Regards > Dirk > > Dirk Moolman > Database Administrator > Reach Technologies > > "People demand freedom of speech as a compensation for the freedom of > thought which they seldom use." > - Kierkegaard >