Oppolzer - Informatik / Blog

B.G. wrote:

Is it possible that we are combining some LUW and z/OS concepts (functions)
together...?    Maybe someone on the LUW side can jump in hear to let us
all know what happens when a dynamic SQL statement is PREPARE in LUW...

I can tell you, right or wrong,  that I have been teaching DB2, writing
about DB2, and presenting about DB2 since the DB2's very beginning, and
have never heard of dynamic SQL doing any less to determine an access path
than static SQL in a DB2 for z/OS environment.

B.



On 7/5/2012 7:41 AM, G.T. wrote:
Hi Terry and Bernd,

I think that most of us who have been with DB2 for a long time (V1.2
in my case) have been told at some time that dynamic binds were
"limited" in he number of possible access paths to evaluate to keep
the process from consuming too much time for a process that could be
a one time use or in a high volume transaction. I beleive we were
told that dynamic would only evaluate the top three access path picks
where a static would evaluate all available paths. This could have
been many releases back that this was told, probably before dynamic
stsement cashing and many other improvements. It is good to hear from
the source that this is no longer true but I wonder if there has been
any study of the cost effectiveness of this?

Thanks,

G.T.