Re: oncheck performance issues

Brian Hampson wrote: > Our oncheck -pe took so long I had to put it in the background.. I used > to get my answer back almost instantaneously w/ 5.x > > B. I did some checking into the differences between 5.x tbcheck -pe and 7.x oncheck -pe and found quite a bit. First off, the 7.x code is much more robust interms of error checking. 5.x code just sort of assumed things worked so that will add some overhead. Second, 7.x oncheck will do 2 for loops from 0 to 0x7FFFFFFF which will take some time even if it only executes 1 statement for most of that loop. 5.x tbcheck -pe did these 2 loops with much smaller range. I think those 2 loops could be coded differently but the reason I would say they changed was do to the fact that you can have many more dbspaces in 7 then 5. Lastly, tbcheck in 5.x was it's own process and it would do it's own work. In 7.x oncheck is a client to the engine and it gets a oncheck thread which has to be run and it passes it's results to oncheck front end via client server fashion, which in 5.x there wasn't this overhead because oncheck/tbcheck just did it itself. Not that any of this will speed up your onchecks, but it could explain it if you are interested. -- ******************************************************************** * Jacques P. Renaut "I'd dazzle you with brilliance * * Informix Advanced Support if I only had the knack..." * * email: jrenaut@informix.com #include <disclamier.h> * ********************************************************************