Re: Minimizing bufwaits

In article <703o1q$89p$1@nnrp1.dejanews.com>, brian@asl.ca () wrote: > In article <vI1++AAsRSJ2EwsS@smooth1.demon.co.uk>, > David Williams <djw@smooth1.demon.co.uk> wrote: > > > >bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress > > seqscans > > >3005610 936 3233912260 2 0 2109 7633098 > > 3122129 > > > > 3122129 sequential scans?? Are you sure that OPTCOMPIND=0 and > > indexes > > are being used where required. > > That's on about 3billion transactions. It's set to OPTCOMPIND=1 > Apparently > indeces are where required. I'm the sysadmin, and the dba isn't much > into > fine tuning...I'm of the feeling, that we can always make things better > :) I believe for performance you fix it until it breaks and back the last change out. But theres little point in tuning Informix to the limit if the server runs like a dog > > > > > To reduce bufwaits increase the number of buffers... > > I guess my concern is that it's possible to OVER increase buffers, but > I'll > have a go at 'er and see what I can do. A gentleman in .nl emailed > today as > well, and has offered some ideas. I'm hoping that there really is room > for > improvement, and that we're not just "getting full" Buffers are like memory you can never have too much (ish) Paul Watson # I don't suffer from WF Software Ltd. # stress, I'm just Tel. (+44) 1436 674729 # a carrier Fax. (+44) 1436 678693 # www.wfsoftware.co.uk