Re: Frequent Update Project: Design Overview of HOT Updates - Mailing list pgsql-hackers

From NikhilS
Subject Re: Frequent Update Project: Design Overview of HOT Updates
Date
Msg-id d3c4af540611101123p671025d2r72651e9a428b4ba5@mail.gmail.com
Whole thread Raw
In response to Re: Frequent Update Project: Design Overview of HOT Updates  ("Pavan Deolasee" <pavan.deolasee@gmail.com>)
List pgsql-hackers
Hi,

On 11/10/06, Pavan Deolasee <pavan.deolasee@gmail.com> wrote:

On 11/10/06, Tom Lane <tgl@sss.pgh.pa.us> wrote:
"Pavan Deolasee" < pavan.deolasee@gmail.com> writes:
> On 11/10/06, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> (2) Isn't this full of race conditions?

> I agree, there  could be race  conditions. But IMO we can handle those.

Doubtless you can prevent races by introducing a bunch of additional
locking.  The question was really directed to how much concurrent
performance is left, once you get done locking it down.


I understand your point and I can clearly see a chance to improve upon the current
locking implementation in the prototype even though we are seeing a good performance
boost for 50 clients and 50 scaling factor with pgbench runs as mentioned by Nikhil.

Regards,
Pavan


Yes, we have done a number of runs with and without autovacuum with parameters like 50 clients, 50 scaling factor and 25000 transactions per client. 50 clients should introduce a decent amount of concurrency. The tps values observed with the HOT update patch (850 tps) were approximately 200+% better than PG82 sources (270).
 
Runs with 25 clients, 25 scaling factor and 25000 transactions produce similar percentage increases with the HOT update patch.
 
Regards,
Nikhils
 
 
 

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

pgsql-hackers by date:

Previous
From: "Simon Riggs"
Date:
Subject: Re: Frequent Update Project: Design Overview ofHOTUpdates
Next
From: NikhilS
Date:
Subject: Re: Frequent Update Project: Design Overview of HOTUpdates