Re: Performance implications of creating many, many sequences - Mailing list pgsql-general

From Rob Sargent
Subject Re: Performance implications of creating many, many sequences
Date
Msg-id 4CC1EB02.60709@gmail.com
Whole thread Raw
In response to Performance implications of creating many, many sequences  (Michael Gardner <gardnermj@gmail.com>)
Responses Re: Performance implications of creating many, many sequences  (Michael Gardner <gardnermj@gmail.com>)
List pgsql-general
Is this "invoice_number" just an id or what might appear an a bill (in
some pretty form etc)?

If the former, just get a unique id over all invoices.  At the very
least it will save time i) in writing where clauses ii) re-creating the
correct id once some one assigns an invoice to the wrong customer.

On 10/22/2010 01:18 PM, Michael Gardner wrote:
> Consider the following table:
>
> CREATE TABLE invoice (
>     account_id integer NOT NULL REFERENCES account,
>     invoice_number integer NOT NULL,
>     UNIQUE (account_id,invoice_number));
>
> I would like to do the equivalent of making invoice_number a serial type, but on a per-account basis. Would it be a
reasonableapproach to create a separate sequence for each individual account? Are there performance implications I
shouldknow about, given that there will be hundreds of thousands of accounts? Is there another approach I should be
lookingat instead? 

pgsql-general by date:

Previous
From: Michael Gardner
Date:
Subject: Performance implications of creating many, many sequences
Next
From: Michael Gardner
Date:
Subject: Re: Performance implications of creating many, many sequences