No, the BCHR is ONLY a measure of the probability that a block will be in the buffer upon re-read:should Buffer Cache Hit ratio be increase or decreasee, in order to improve the performance?
Meaningless, no, super-useful, no.BCHR is a meaningless indicator for the performance of the database.
That's nonsense. He did not say that anything was broken.Stop fighting symptoms, fix the application.
This, I believe.Well-tuned systems are an exception in my book . . .
YOU DO NOT HAVE ANY verifiable experience whatsoever. . . .In my experience tuning the database
Nope, totally untrue, and there are reproduceable benchmarks to prove it:purchasing solid state disks won't help, EVER.
May I ask you a question? Obviously, you can flood the buffer with reads against pre-cached rows and change the BCHR at-will, but does that mean that we conclude that it's totally useless?For those individuals everything taking seriously the first part was a joke.
May I ask you a question? Obviously, you can floodDon, You know I don't want to beat dead horse again, again and again. Beating dead horses is not one of my hobbies. You showed OP your article(-s) I showed him an article explaining a particular nature of this ratio ;) and let's allow the OP make his decision.
the buffer with reads against pre-cached rows and
change the BCHR at-will, but does that mean that we
conclude that it's totally useless?
I find some limited use in monitoring BCHR, to alert
me to load changes, something unexpected going on.