Re: [WIP PATCH] for Performance Improvement in Buffer Management

On Monday, November 19, 2012 5:53 AM Jeff Janes wrote:
On Sun, Oct 21, 2012 at 12:59 AM, Amit kapila <amit.kapila@huawei.com> wrote:
> On Saturday, October 20, 2012 11:03 PM Jeff Janes wrote:
>
>>Run the modes in reciprocating order?
>> Sorry, I didn't understood this, What do you mean by modes in reciprocating order?

> Sorry for the long delay.  In your scripts, it looks like you always
> run the unpatched first, and then the patched second.

   Yes, thats true.

> By reciprocating, I mean to run them in the reverse order, or in random order.

Today for some configurations, I have ran by reciprocating the order.
Below are readings:
Configuration
16GB (Database) -7GB (Shared Buffers)

Here i had run in following order
        1. Run perf report with patch for 32 client
        2. Run perf report without patch for 32 client
        3. Run perf report with patch for 16 client
        4. Run perf report without patch for 16 client

Each execution is 5 minutes,
    16 client /16 thread    |   32 client /32 thread
   @mv-free-lst @9.3devl    |  @mv-free-lst @9.3devl
-------------------------------------------------------
      3669            4056            |   5356            5258
      3987            4121            |   4625            5185
      4840            4574            |   4502            6796
      6465            6932            |   4558            8233
      6966            7222            |   4955            8237
      7551            7219            |   9115            8269
      8315            7168            |   43171            8340
      9102            7136            |   57920            8349
-------------------------------------------------------
      6362            6054            |   16775            7333