> > mem_max  64m
>
> If total amount of documents in memory reach mem_max, then we start drop
> documents without save to disk.
>
> > lo_mark  18m
>
> When total amount of documents in memory reach lo_mark we start swpa thm
out
> to disk. this process swap out as much documents as we need to return back
> to lo_mark (and a bit more, so we will no need to swap them out too
> frequently)
>

So is it wish to increase lo_mark to over 30mb or ever over 8mb?


> > What results would occur if those values would have been increase in
> > relation to others especially "db_cache_mem"?
>
> if you increase db_cache_mem you will probably see less(and maybe bursty)
> disk io with database.
>

Understood.

Edward


=====================================================================
If you would like to unsubscribe from this list send message to
[EMAIL PROTECTED] with "unsubscribe oops-eng" in message body.
Archive is accessible on http://lists.paco.net/oops-eng/

Дати відповідь електронним листом