Re: CPU % Util spikes during single insert

From: <michael_bialik_at_my-deja.com>
Date: Sat, 15 Jan 2000 08:39:41 GMT
Message-ID: <85pboc$cq6$1@nnrp1.deja.com>

Hi.

Are you inserting VERY long rog into your table ( you specified a number of VARCHAR2(1024) fields in your table )? If the answer is yes then check what happens on insert of short row ( one letter only for each varchar2 field ). Check your space allocations for the table ( INITIAL/NEXT extents, PCTFREE/PCTUSED etc. ).
Another possibility is to execute the INSERT statement with SQL TRACE set nad to analyze TKPROF output ( look for the number of recursive statements ).

HTH. Michael.

In article <387FB6CE.41027C5C_at_acad.utah.edu>, smcmurray_at_acad.utah.edu wrote:
> I'm experiencing a problem with an application where selects and
inserts
> work fine on all tables except one.
>
> Inserting to a similar table is instant, but inserting a single record
> to the problem table causes CPU %Util to spike from 0% to 100%; Mem %
> Util jumps from 0% to 7.4%; and it takes 4+ minutes to complete the
> insert!
>
> The table structure is:
>
> NUMBER
> VARCHAR2(8)
> VARCHAR2(32)
> VARCHAR2(32)
> VARCHAR2(32)
> DATE
> VARCHAR2(128)
> VARCHAR2(128)
> VARCHAR2(128)
> VARCHAR2(64)
> VARCHAR2(2)
> VARCHAR2(9)
> VARCHAR2(64)
> VARCHAR2(64)
> VARCHAR2(24)
> VARCHAR2(24)
> VARCHAR2(128)
> VARCHAR2(1)
> VARCHAR2(1024)
> VARCHAR2(1024)
>
> There are <10,000 records in the table.
>
> Inserting into a table with an even larger structure is instant.
> Inserting into tables with 250,000+ records is instant.
>
> Any hints on what the problem might be?
>
> --
>
> Sean McMurray
>
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Sat Jan 15 2000 - 02:39:41 CST