Read ahead & Buffer wait ratio query.

Hello all, Below is the formulla I use for calculating the Read Ahead usage : ((ixdaRA + idxRA + daRA) / RApgsused) * 100 Many I times I see the percentage is above 100. Does it mean that the page which was read ahead was read/used twice or more. I have seen this formulla in Informix list mails for calculating Buffer wait ratio or Buffer turnover ratio : ((bufwaits/(pagreads + bufwrits)) * 100) Can anybody please explain me the mean of this in simple english? Thanks a lot for all your replies. _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com

In article <9573vc$erl$1@news.xmission.com>, "Tony Woods" <twoods786@hotmail.com> wrote: > > Hello all, > > Below is the formulla I use for calculating the Read Ahead usage : > ((ixdaRA + idxRA + daRA) / RApgsused) * 100 > Many I times I see the percentage is above 100. Does it mean that the page > which was read ahead was read/used twice or more. > Read ahead usage = (RApgused/(ixdaRA + idxRA + daRA))*100 > I have seen this formulla in Informix list mails for calculating Buffer wait > ratio or Buffer turnover ratio : > ((bufwaits/(pagreads + bufwrits)) * 100) > Can anybody please explain me the mean of this in simple english? Buffer wait ratio (Number of times you waited on a buffer)/(the number of times that a buffer was modified) This is an attempt to identify contention for buffers(and maybe LRU's). Art Kagel tends to give good information on this topic, you might search the cdi archives for one of his postings on this topic, or see if he responds to this post. Buffer turn over ratio is something like (pagereads/BUFFERS)/(time since statistics last initialized) This gives a fair indication of how long a page stays cached after it is read. Hope this answers your questions, Will Sent via Deja.com http://www.deja.com/