mkambol at gmail.com
Mon Feb 8 11:13:08 EST 2010
Thanks for the background, Julian. We are using the Cache function
explicitly in a set of queries that do some statistical calculations, where
a part of the calculation has the potential to be repeated 10s of thousands
of times, and we were curious why Cache function was not included in the
On Fri, Feb 5, 2010 at 2:13 PM, Julian Hyde <jhyde at pentaho.com> wrote:
> I presume because Cache is introduced as a physical optimization and the
> query has the same semantics if it is not present.
> The planner introduces Cache in certain circumstances such as directly
> under a Rank function. Maybe the planner is not smart enough to not
> introduce it twice.
> Because of these issues, I'd rather that these kind of optimizations didn't
> modify the MDX parse tree at all. In other words, Cache shouldn't be a
> function, it should be a Calc. But it is what it is.
> *From:* mondrian-bounces at pentaho.org [mailto:mondrian-bounces at pentaho.org]
> *On Behalf Of *Matt Campbell
> *Sent:* Thursday, February 04, 2010 12:25 PM
> *To:* Mondrian developer mailing list
> *Subject:* [Mondrian] CacheFunDef.unparse()
> CacheFunDef has an override for unparse which has the effect of suppressing
> the function from the unparsed output. Does anyone know the rationale for
> this behavior?
> Mondrian mailing list
> Mondrian at pentaho.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Mondrian