Re: Performance problems

You know,
It'd be nice if there was a system table, or command that showed the number and/or percentage of tuples needing to be vacuumed. If table are only affected by tuples formerly in them, then the table/function could show the value per table. If any discarded tuples affect all tables, then a global table/function would be warranted.

A minimally compute intensive chron job or ON DELETE trigger could then call vacuum full at a certain percentage. Also, what kind of memory management ( in the postgres application ) could be written that pushes deleted/unused tuples out of any caches or scopes to at least lengthen the time between vacuums?

Shridhar Daithankar wrote:
> On Friday 25 April 2003 20:23, marco wrote:
>
>>I unfortunately don't understand the whole thing totally, but if I dump
>>the database (with pg_dump), delete it and restore it, the time values
>>for reading and writing have decreased to a normal level and begin to
>>increase again.
>>
>>It seems to me, that I do sth. fundamentally wrong :( But even after
>>searching google and the postgresql archives I don't see the light at
>>all...
>
>
> You need to vacuum full everytime you delete large amount of rows and vacuum
> analyze every time you insert/update large amount of rows..
>
> I would say large amount==50K row is a good start.. So after 20 runs of tool,
> run vacuum once.. Try it and let us know..
>
> Shridhar
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
> http://archives.postgresql.org
>