long checkpoint duration

Hi,   The age old question I guess. But our checkpoint durations are consistently near 20seconds. ids10.0 fc5 on ibm p570 running aix 5.3 7 cpus 41g ram. ckptintvl is 300, I am considering halving that. We have 128 lru's and 128 cleaners Art's ratio tool shows all good except a high btr. We have 2557440 buffers, but I am pretty sure if we add much more we'll start paging. Anyway , that probably isn't affecting the checkpoint. onstat -g iov shows the inverted pyramid with 7 kio threads. Aside from worrying about fragmentation and data distribution issues in our FastT900 san, is there anything I'm missing that I should look at real quick to reduce the chkpoint interval ? Thanks, Floyd

"Floyd Wellershaus" <floyd@fwellers.com> wrote in message news:mailman.462.1202836569.20610.informix-list@iiug.org... Hi, The age old question I guess. But our checkpoint durations are consistently near 20seconds. ids10.0 fc5 on ibm p570 running aix 5.3 7 cpus 41g ram. ckptintvl is 300, I am considering halving that. We have 128 lru's and 128 cleaners Art's ratio tool shows all good except a high btr. We have 2557440 buffers, but I am pretty sure if we add much more we'll start paging. Anyway , that probably isn't affecting the checkpoint. onstat -g iov shows the inverted pyramid with 7 kio threads. Aside from worrying about fragmentation and data distribution issues in our FastT900 san, is there anything I'm missing that I should look at real quick to reduce the chkpoint interval ? What does sar -d, say sar -d 60 15, tell you about the disk service times?

How many partitions are being accessed? Lots of partitions will cause the checkpoint to slow because they have to be flushed out. Have you thought about moving to Cheetah? The non-blocking checkpoint for heavy oltp systems will remove this problem. Customers moving to Cheetah have experience anything from 50% to 500% performance improvement with Cheetah when the LRU's are made to be less aggressive. On Feb 12, 8:18 pm, "Neil Truby" <neil.tr...@ardenta.com> wrote: > "Floyd Wellershaus" <fl...@fwellers.com> wrote in message > > news:mailman.462.1202836569.20610.informix-list@iiug.org... > Hi, > The age old question I guess. But our checkpoint durations are > consistently near 20seconds. > > ids10.0 fc5 on ibm p570 running aix 5.3 7 cpus 41g ram. > ckptintvl is 300, I am considering halving that. > We have 128 lru's and 128 cleaners > Art's ratio tool shows all good except a high btr. > We have 2557440 buffers, but I am pretty sure if we add much more we'll > start paging. Anyway , that probably isn't affecting the checkpoint. > > onstat -g iov shows the inverted pyramid with 7 kio threads.> > Aside from worrying about fragmentation and data distribution issues in our > FastT900 san, is there anything I'm missing that I should look at real quick > to reduce the chkpoint interval ? > > What does sar -d, say sar -d 60 15, tell you about the disk service times?