[Info-Ingres] Cache hit ratio 20% ummm is that good?
Roy Hann
specially at processed.almost.meat
Thu Nov 15 04:32:30 CST 2007
"Paul White" <shift7solutions at gmail.com> wrote in message
news:mailman.65.1195102916.18164.info-ingres at kettleriverconsulting.com...
>I was horrified to find a cache hit ratio of under 20% on one of my
>customer
> servers recently. Upon investigation the story is more interesting...
>
> Fix Calls 169880982
> Hits 28015058
> I run a trace point DM420 every hour to collect useful stats on the cache.
> Somewhere between 10am and 11am the FIX CALLS reset back to zero. Then
> later
> the next day the HITS reset back to zero. It seems 2^32 is the maximum.
I've not observed that myself. What version is this? What does IMA say?
[snip]
> Have I bored you all to tears?
Nope. I can truthfully say this is one of the three most interesting post
this week.
> The server is rebooted monthly so the
> problem only shows up after some very heavy processing (eg a process went
> crazy)
>
> ps.. Would anyone like to comment on my GREAD numbers? I think they are
> too
> high for a transaction based database.
GREADs are a pain. At some point in the recent past, some over-eager soul
decided to change that counter to indicate the number of pages read by group
reading, rather than the number of group read operations. I never remeber
which version the change was introduced, and I am not sure I remember
whether it has been reversed or not, and if so, in which version it was
reversed.
Roy
More information about the Info-Ingres
mailing list