Re: Btree Index on PostgreSQL and Wiredtiger (MongoDB3.2) - Mailing list pgsql-hackers

From Jeff Janes
Subject Re: Btree Index on PostgreSQL and Wiredtiger (MongoDB3.2)
Date
Msg-id CAMkU=1w0BHPgJ65uZ=TBGkUJ=h-61vEcyLN3Etk6qJKEiEWaGQ@mail.gmail.com
Whole thread Raw
In response to Re: Btree Index on PostgreSQL and Wiredtiger (MongoDB3.2)  (Mark Kirkwood <mark.kirkwood@catalyst.net.nz>)
Responses Re: Btree Index on PostgreSQL and Wiredtiger (MongoDB3.2)  (Andrew Gierth <andrew@tao11.riddles.org.uk>)
Re: Btree Index on PostgreSQL and Wiredtiger (MongoDB3.2)  (Mark Kirkwood <mark.kirkwood@catalyst.net.nz>)
List pgsql-hackers
On Fri, Aug 12, 2016 at 1:40 AM, Mark Kirkwood
<mark.kirkwood@catalyst.net.nz> wrote:
> After examining the benchmark design - I see we are probably not being
> helped by the repeated insertion of keys all of form 'userxxxxxxx' leading
> to some page splitting.

But shouldn't that still leave us with a 75% full index, rather than
slightly over 50% full?

The leaf pages start at 50%, grow to 100%, then split back to 50%,
then grow back to 100%.  So the average should be about 75%.


> However your index rebuild gets you from 5 to 3 GB - does that really help
> performance significantly?

It can make a big difference, depending on how much RAM you have.

Cheers,

Jeff



pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: TODO item: Implement Boyer-Moore searching in LIKE queries
Next
From: Robert Haas
Date:
Subject: Re: Add hint for function named "is"