Re: MVCC for massively parallel inserts - Mailing list pgsql-general

From Tom Lane
Subject Re: MVCC for massively parallel inserts
Date
Msg-id 6371.1073355961@sss.pgh.pa.us
Whole thread Raw
In response to MVCC for massively parallel inserts  (Steven D.Arnold <stevena@neosynapse.net>)
Responses Re: MVCC for massively parallel inserts  ("Joshua D. Drake" <jd@commandprompt.com>)
List pgsql-general
"Steven D.Arnold" <stevena@neosynapse.net> writes:
> So I really need many of the machines in the cluster to be
> able to simultaneously insert.  Is this practicable in a clustered
> environment for Postgres?

No.

You are mistaken to think that CPU is the bottleneck, anyway.  It is
going to be disk bandwidth that's your problem, and throwing more CPUs
at it would not help.  If your insert load exceeds the available disk
bandwidth, I think you have little choice but to figure out a way to
divide the workload among multiple independent databases.

            regards, tom lane

pgsql-general by date:

Previous
From: Steven D.Arnold
Date:
Subject: MVCC for massively parallel inserts
Next
From: "Joshua D. Drake"
Date:
Subject: Re: MVCC for massively parallel inserts