Re: ANALYZE sampling is too good - Mailing list pgsql-hackers

From Greg Stark
Subject Re: ANALYZE sampling is too good
Date
Msg-id CAM-w4HPdMtPiS9A3cj1gfO0vjUT4eK3KXyKpv8FLrnv1cN529w@mail.gmail.com
Whole thread Raw
In response to Re: ANALYZE sampling is too good  (Albe Laurenz <laurenz.albe@wien.gv.at>)
Responses Re: ANALYZE sampling is too good
Re: ANALYZE sampling is too good
List pgsql-hackers
<p dir="ltr"><br /> On 10 Dec 2013 08:28, "Albe Laurenz" <<a
href="mailto:laurenz.albe@wien.gv.at">laurenz.albe@wien.gv.at</a>>wrote:<br /> ><br /> ><br /> > Doesn't
allthat assume a normally distributed random variable?<br /><p dir="ltr">I don't think so because of the law of large
numbers.If you have a large population and sample it the sample behaves like a normal distribution when if the
distributionof the population isn't. 

pgsql-hackers by date:

Previous
From: Claudio Freire
Date:
Subject: Re: Optimize kernel readahead using buffer access strategy
Next
From: Alvaro Herrera
Date:
Subject: Re: New option for pg_basebackup, to specify a different directory for pg_xlog