Re: MVCC cons

From: Gregory Stark <stark(at)enterprisedb(dot)com>
To: "Kenneth Downs" <ken(at)secdat(dot)com>
Cc: "RPK" <rpk(dot)general(at)gmail(dot)com>, <pgsql-general(at)postgresql(dot)org>
Subject: Re: MVCC cons
Date: 2007-08-14 22:20:50
Message-ID: 87bqd9oj0d.fsf@oxford.xeocode.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

"Kenneth Downs" <ken(at)secdat(dot)com> writes:

> RPK wrote:
>> I want to know whether MVCC has cons also. Is it heavy on resources? How
>> PGSQL MVCC relates with SQL Server 2005 new Snapshot Isolation.

Well the fundamental con of MVCC versus serializing everything using locks is
that you have to keep more data around which takes space and i/o resources.

> Speaking as an end-user, I can give only one I've ever seen, which is
> performance. Because of MVCC, Postgres's write performance (insert and update)
> appears on my systems to be almost exactly linear to row size. Inserting 1000
> rows into a table with row size 100 characters takes twice as long as inserting
> 1000 rows into a table with row size 50 characters.

Well, that's going to be true (assuming i/o is your limiting factor at all)
regardless of your architecture. There's no magic here, more bytes take more
blocks of space which take more time to write or read.

--
Gregory Stark
EnterpriseDB http://www.enterprisedb.com

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2007-08-14 23:50:28 Re: pqlib in c++: PQconnectStart PQconnectPoll
Previous Message Lew 2007-08-14 22:16:16 Re: Compound Indexes