Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Poor Buffer Hit Ratio

Re: Poor Buffer Hit Ratio

From: <junderhi_at_my-deja.com>
Date: 2000/05/16
Message-ID: <8frlkt$6cv$1@nnrp1.deja.com>#1/1

This is what I get for output from your provided queries. Looks like you were right on.

DBWR Free Buffers Found 4405170
DBWR Make Free Requests 3020
Avg. No. Free Buffers 1458

PL/SQL procedure successfully completed.

Elapsed: 00:00:00.14

                               Get Hit            IGet Hit
NAME                             Ratio     SLEEPS    Ratio
------------------------------ ------- ---------- --------
cache buffers chains              99.2       2287    100.0
cache buffer handles             100.0          0      0.0
cache buffers lru chain           99.8         35     99.9


In article <8esdtp$f6c$1_at_nnrp1.deja.com>,   Mark D Powell <markp7832_at_my-deja.com> wrote:
> In my opinion you probably have more buffers than you effecively use:
> The buffer pool hit ratio is unimportant in itself. What counts is
 how
> well your application runs. If it is running well a hit ratio of 70%
 is
> great, and if it is running slow then a hit ratio of 90% is not good
> enough. How many free buffers do you have on average and what are
 your
> buffer pool latch hit ratios? These will give you a better indication
> of how effecively application is using the buffer pool. I believe
 that
> the OS and your Oracle shadow processes could make better use of most
> most of the memory your are allocating to the buffer pool. Here is an
> Oracle note that states that for ver 7.3.4 and 8 that the statistics
> used to calculate the buffer pool hit ratio are invalid to begin with.
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Tue May 16 2000 - 00:00:00 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US