On Thu, 4 May 2000, Tom Lane wrote:
> The Hermit Hacker <scrappy@hub.org> writes:
> >> ! system usage stats:
> >> ! 0.436792 elapsed 0.275139 user 0.157033 system sec
> >> ! [0.283135 user 0.173026 sys total]
> >> ! 0/0 [0/0] filesystem blocks in/out
> >> ! 0/149 [0/332] page faults/reclaims, 0 [0] swaps
> >> ! 0 [0] signals rcvd, 0/0 [2/2] messages rcvd/sent
> >> ! 0/8 [2/9] voluntary/involuntary context switches
>
> > How is this to be read? I'm looking at it, and reading it as:
>
> > 0 - voluntary
> > 8 - involuntary
>
> > But what about the [2/9]?
>
> I believe the numbers outside brackets are for the particular query
> cycle, and the ones in brackets are total for the process (ie, total
> since backend start).
>
> I didn't design the printout format ;-) ... not real sure what the
> point is of printing the total-since-start numbers ...
Okay, that explains that :)
Now, Mitch's results for v7.0 showed something like:
0/12 filesystem blocks in/out
You intepreted that as 12 reads from the file system ... 'out' I would
have interpreted as writes to the file system, which made zero sense
... do we have our 'in/out's backwards here?
One thing that would be nice (who wrote these stats?) would be some way to
be able to determine a suitable setting for -S from this ... someway to
know that an ORDER BY needed to swap to disk because it needed 32Meg when
only 16Meg was allocated for it ... would help give an indication where
performance could be improved by either just raising -S (in Mitch's case,
where lots of RAM is available) or more RAM should be added ...
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org