Chris Hughes
ProgressTalk.com Sponsor
I've seen this question asked about every database platform I've ever worked on without a sensible solution but want to check my understanding is correct here for OpenEdge.
on 10.2b+ I have a 4GB database, assume my -B is 1GB and my -B2 is 2GB (LRU not activated).
If I start this DB, then immediately run a proutil dbanalys, every block gets read.
Therfore my B2 is populated as I don't fill it, my B has the last 1GB of data I read.
Therefore users don't need to read any B2 from disk and although my B is full they stand a better chance of grabbing a cached block than if I hadn't ran the analysis.
Promon seems to support my theory just looking for some confirmation by the experts
Thanks in advance
on 10.2b+ I have a 4GB database, assume my -B is 1GB and my -B2 is 2GB (LRU not activated).
If I start this DB, then immediately run a proutil dbanalys, every block gets read.
Therfore my B2 is populated as I don't fill it, my B has the last 1GB of data I read.
Therefore users don't need to read any B2 from disk and although my B is full they stand a better chance of grabbing a cached block than if I hadn't ran the analysis.
Promon seems to support my theory just looking for some confirmation by the experts
Thanks in advance