Showing posts with label olap. Show all posts
Showing posts with label olap. Show all posts

Wednesday, February 15, 2012

EDST impact over the OLAP cubes.

Hi Everyone,

As the EDST time is going to start sometimes around next time. Would like to know if this could affect to the OLAP Cubes?

Thanks,

If you have any kind of time-related calculations in your cubes, these calculations should be taking into account DST. You might want to review your cube design to see if you need to tweak the design a little bit.

Edward.
--
This posting is provided "AS IS" with no warranties, and confers no rights.

|||

I had checked with our microsoft account manager and he told me that there was no need to apply any DST patches for any SQL applications except Notification services. However we will need to apply the DST patch at the OS level.

Edward: If the patch is applied at the OS level, Analysis Services should not need anything else. This issue is big at our company and I just want to confirm again with you.

Thanks for all your help!!

Rgds
Hari

Edition Question for Analysis Services

Hi All

I come with Oracle background and we are planning to use SQL Server 2000 Analysis Services for OLAP.

Brief background on our app: The largest of our fact tables has about 500,000 rows per month. We plan to store three years worth of data. We will have other smaller cubes that we will have to build in addition to this. The source data will come from Oracle tables.

I am struggling to decide between the two SQL Server editions (Standard and Enterprise). I'm highlighting the things that I need some help with in order to make the decision:

Memory Limitation
------
Standard Edition only supports up to 2GB. Considering relatively large cube size (is it?), is this a limitation in our case?

Calcualted Cells (only in Enterprise Edition)
------
I didn't fully understand this one in the documentation, not having spent a lot of time. We do have a number of calculations such as the Revenue/Expense, Margin calculations, etc. Is this an important feature that I should consider?

Partitions:
----
I doubt that we will be partitioning across multiple servers (may not need distributed partioning, linked cubes). However, within a single database/instance, do I have the option to partition the cube (does it help to do so for cube manageability) if I went with Standard Edition? Also, given the large fact tables, I may want to incrementally add monthly data to the cube. Would it help to partition the cube in this case and, do I need Enterprise Edition for this?

HTTP Internet Support: (Enterprise Edition only)
-------
I would think that the HTTP access is applicable to intranet applications (ours is one). Does this support in Enterprise Edition automatically means that the data is available for analysis over the web? What tool would I use to access the data (other than the third party tools such as Crystal Analysis, Brio, etc)? I am trying to see what this means and how this relates to my decision on Std Vs Enterprise Editions.

I really appreciate any input you may have on this.

Thanks in advance
Murthy JarugumilliI would suggest to adopt Enterprise Edition as you already know the limitaitons of Standard edition.

In general Standard edition is used as a database server for a small workgroup or department.