Re: Storing database in cluster (Memory)

roopa perumalraja wrote:
> Hi all,
>
> As I am inserting 100million rows daily into partitioned tables
> (daily wise), it is getting slower.

What is - the inserts? By how much? What tables? What indexes? How are
you inserting these rows?

> Even the retrivel of data, select
> statement on those tables takes about 30 mintues.

All selects take 30 minutes, regardless what work they do? Or do you
have specific selects that are causing problems?

> I have tried
> increasing the parameters in postgres.conf but still that doesn't
> help me much as the no of rows are huge.

What parameters, to what values? What hardware are you running on? What
load do you place on it?

> Will the idea of storing the
> database in cluster (memory) increase the performance of
> insert/update/select in the table in a suitation like this? Thanks a
> lot in advance.

A better question would be "Why isn't my data being cached?". The
operating-system should cache regularly used files.

Still - more details please roopa and we'll see if anyone can help you.

--
Richard Huxton
Archonet Ltd