Re: 110,000,000 rows

> -----Original Message-----
> From: pgsql-general-owner(at)postgresql(dot)org [mailto:pgsql-general-
> owner(at)postgresql(dot)org] On Behalf Of Thom Brown
> Sent: Wednesday, May 26, 2010 2:14 PM
> To: John Gage
> Cc: PostgreSQL - General
> Subject: Re: [GENERAL] 110,000,000 rows
>
> On 26 May 2010 21:29, John Gage <jsmgage(at)numericable(dot)fr> wrote:
> > Please forgive this intrusion, and please ignore it, but how many
> > applications out there have 110,000,000 row tables?  I recently
> multiplied
> > 85,000 by 1,400 and said now way Jose.
> >
> > Thanks,
> >
> > John Gage
> >
>
> There's no reason why it can't have that many rows. There's no limit
> on the number of rows for tables, only table sizes, which is limited
> to 32TB. If you ever have a table that big though, you probably need
> to rethink your schema.

There really are domains that big, so that there is no more normalization or other processes to mitigate the problem.

Examples:
Microsoft's registered customers database (all MS products bought by any customer, including operating systems)
Tolls taken on the New Jersey road system for FY 2009
DNA data from the Human Genome Project
Protein data from the Protein Folding Project
The US Census bureau's Tiger/Line data
Online orders processed by Amazon.com
Cellular phone calls for t-Mobile for 2008
FedEx shipments worldwide in 2008

We work with mainframe data on a regular basis and files of that size are not really very unusual. Expansion of data is exponential over time. We need to prepare for it. Database systems that cannot handle the volume will be supplanted by those that can.