EXPLAIN - Influencing the Optimizer

[Bruce Pullen]

DB2 UDB 7.2 on AIX5.2

I'm running (VISUAL) EXPLAIN from DB2 Control Center on various SQL
statements, but am restricted to using unrealistically small data
volumes. I've heard that it is possible to kid the EXPLAIN into
thinking it's dealing with much larger data volumes. Assuming that
this would be productive, can anyone tell me how I can do this?

Many thanks
Bruce.

[Mark A]

"Bruce Pullen" <br**********@hotmail.com> wrote in message
news:9d**************************@posting.google.c om...

DB2 UDB 7.2 on AIX5.2

I'm running (VISUAL) EXPLAIN from DB2 Control Center on various SQL
statements, but am restricted to using unrealistically small data
volumes. I've heard that it is possible to kid the EXPLAIN into
thinking it's dealing with much larger data volumes. Assuming that
this would be productive, can anyone tell me how I can do this?

Many thanks
Bruce.

You can update the statistics in the catalog. The most important in your
case is probably the card (number of rows) on the SYSCAT.TABLES.

[PM \(pm3iinc-nospam\)]

http://publib.boulder.ibm.com/infoce...help/index.jsp
SEARCH : DB2LOOK

WIth the -m option (in combination with others), you should be able to work
it out.
Generate the DDL file, modify it, run it, explain again.

PM

"Bruce Pullen" <br**********@hotmail.com> a écrit dans le message de
news:9d**************************@posting.google.c om...

DB2 UDB 7.2 on AIX5.2

I'm running (VISUAL) EXPLAIN from DB2 Control Center on various SQL
statements, but am restricted to using unrealistically small data
volumes. I've heard that it is possible to kid the EXPLAIN into
thinking it's dealing with much larger data volumes. Assuming that
this would be productive, can anyone tell me how I can do this?

Many thanks
Bruce.

[Bruce Pullen]

Thank you both for your help. Perfect!