Hi all,

I have a small cube. The total size of the cube is 16.14
MB and the total size of the database is 21.06 MB.

It has 17 dimensions (2 of which are virtual). It has 10
partitions. I have 15 measures plus 6 calculated members.

My calculated members take a while to come up. I looked at
the performance counters and found that 97% of the queries
are being answered from Files. Only the remaining 3% are
answered dirently from cache. Since my cube is so small
can I cache the entire cube. ? I have 500 MB of RAM, not
sure how much Cache I have.

How can do this so that most of my queries are run from
the cache? Is there a parameter or something that I need
to set or is it more involved? Any ideas would be
appreciated. Also, if you can point me to some good
documentation on this topic.

Thanks a lot,
Amir

Try this link, lots of info:

http://www.microsoft.com/technet/tre...e/ANSvcsPG.asp

If you are not using MOLAP, use it, and depending on how much your users
drill into your cube, you want to use at least 90% for the storage design
and pre-aggregating the aggs. You can do what is called "warm the query
results cache" by looking at your AS query log and determine what dims are
most commonly hit and running queries that look at those dims to load into
the results cache, this is detailed in the above URL.

HTH

--
Ray Higdon MCSE, MCDBA, CCNA
--
"Amir" <acharania (AT) trimontrea (DOT) com> wrote