On Wed, Apr 15, 2015 at 10:07 AM, Robert Haas <robertmh...@gmail.com> wrote: > > On Wed, Apr 15, 2015 at 12:15 AM, Amit Kapila <amit.kapil...@gmail.com> wrote: > > IIUC, this will allow us to increase usage count only when the buffer > > is touched by clocksweep to decrement the usage count. > > Yes. > > > I think such a solution will be good for the cases when many evictions > > needs to be performed to satisfy the workload, OTOH when there are > > not too many evictions that needs to be done, in such a case some of > > the buffers that are accessed much more will have equal probability to > > get evicted as compare to buffers which are less accessed. > > Possibly, but I think it's even worse under the current algorithm. > Under this proposal, if we go for a long time without any buffer > evictions, every buffer usage's count will top out at 2 more than > wherever it was after the last clock sweep. In the worst case, every > buffer (or most of them) could end up with the same usage count. But > under the status quo, they'll all go to 5, which is an even bigger > loss of information, and which will make the first eviction much more > expensive than if they are all pegged at 2 or 3. >
Okay, got your point. On further thinking on this idea, it occurred to me that with this idea you are trying to reduce the clock-sweep time which in-turn will inturn improve the chances of finding usable buffer in less time under high pressure, if that is true then I think we should have seen the similar gain even with the bgreclaimer idea which I have tried sometime back, because that has also reduced the time for clock-sweep. > There could be ways to improve things further, such as by slowly > advancing the clock sweep even when no eviction is required. But > that's a bit tricky to do, too. You'd like (perhaps) to advance it > one step for every buffer allocation, but you don't want a centralized > counter, or unnecessary contention on the clock sweep when no eviction > is necessary in the first place. There's probably some way to make it > work, though, if we put our mind to it. I'm inclined to try the > simpler approach first and see what it buys. > Sure, I think it is worth a try. With Regards, Amit Kapila. EnterpriseDB: http://www.enterprisedb.com
