Re: scalablility problem

From: Xiaoning Ding
Subject: Re: scalablility problem
Date: ,
Msg-id: 460D8371.7080405@cse.ohio-state.edu
(view: Whole thread, Raw)
In response to: Re: scalablility problem  (Tom Lane)
Responses: Re: scalablility problem  (Scott Marlowe)
Re: scalablility problem  (Ron Mayer)
List: pgsql-performance

Tree view

scalablility problem  (Xiaoning Ding <-state.edu>, )
 Re: scalablility problem  ("Dave Dutcher", )
 Re: scalablility problem  (Tom Lane, )
  Re: scalablility problem  (Xiaoning Ding <-state.edu>, )
   Re: scalablility problem  (Scott Marlowe, )
   Re: scalablility problem  (Ron Mayer, )
    Re: scalablility problem  (Xiaoning Ding <-state.edu>, )
     Re: scalablility problem  (Dave Cramer, )
 Re: scalablility problem  (Scott Marlowe, )
  Re: scalablility problem  (Xiaoning Ding <-state.edu>, )
   Re: scalablility problem  ("Joshua D. Drake", )
    Re: scalablility problem  (Tom Lane, )
     Re: scalablility problem  (Xiaoning Ding <-state.edu>, )
 Re: scalablility problem  (Christopher Browne, )
  Re: scalablility problem  (Guido Neitzer, )
   Re: scalablility problem  (Michael Stone, )
    Re: scalablility problem  (Xiaoning Ding <-state.edu>, )
  Re: scalablility problem  (Xiaoning Ding <-state.edu>, )

Thanks guys,

I update PG to 8.2.3. The results are much better now.
1 process :    0.94 second
2 processes: 1.32 seconds
4 processes: 2.03 seconds
8 processes: 2.54 seconds

Do you think they are good enough?
BTW where can I found some info on what 8.2.3 did to improve
scalability compared with pre 8.1 versions?


Xiaoning

Tom Lane wrote:
> Xiaoning Ding <-state.edu> writes:
>> When I run multiple TPC-H queries (DBT3) on  postgresql, I found the system
>> is not scalable.  My machine has 8GB memory, and 4 Xeon Dual Core processor
>> ( 8 cores in total). OS kernel is linux 2.6.9. Postgresql is 7.3.18.
>
> If you are not running PG 8.1 or later, it's really not worth your time
> to test this.  Multiprocessor scalability was hardly even on the radar
> in 7.3 days.
>
>             regards, tom lane
>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: explain analyze is your friend
>
>



pgsql-performance by date:

From: Christopher Browne
Date:
Subject: Re: scalablility problem
From: Guido Neitzer
Date:
Subject: Re: scalablility problem