Re: Frequently updated tables - Mailing list pgsql-hackers

From pgsql@mohawksoft.com
Subject Re: Frequently updated tables
Date
Msg-id 19034.24.91.171.78.1086736605.squirrel@mail.mohawksoft.com
Whole thread Raw
In response to Re: Frequently updated tables  (Mark Kirkwood <markir@coretech.co.nz>)
Responses Re: Frequently updated tables  (Mark Kirkwood <markir@coretech.co.nz>)
Re: Frequently updated tables  ("Jim C. Nasby" <decibel@decibel.org>)
List pgsql-hackers
>
>
> pgsql@mohawksoft.com wrote:
>
>>I've been down several roads about how to handle data that has to change
>>on a very frequent and rapid manner.
>>
>>Think about summary tables, WEB session tables, etc. As great as MVCC is
>>for the vast majority of uses. The overhead of updates and deletes can
>>kill a project that needs to constantly update tables.
>>
>>
>>
>>
>
> Are you saying that MVCC has *by design* a higher overhead for updates
> and deletes? or are you referring to the gradual loss of performance as
> a consequence of many dead tuples?
>
> I am guessing you mean the latter, but best to be sure :-)
The best phrasing would be "the accumulating overhead of deletes and
updates."

Yes.


pgsql-hackers by date:

Previous
From: Mark Kirkwood
Date:
Subject: Re: Frequently updated tables
Next
From: Alvaro Herrera
Date:
Subject: Re: cvs head : broken regression tests ?