dbTalk Databases Forums  

OLAP design question

comp.databases.olap comp.databases.olap


Discuss OLAP design question in the comp.databases.olap forum.



Reply
 
Thread Tools Display Modes
  #1  
Old   
csharp_c0d3r
 
Posts: n/a

Default OLAP design question - 09-16-2004 , 06:22 PM






To all OLAP gurus:

My organization is considering an OLAP solution for business
intelligence/reporting. I think this is great but I have serious
misgivings about the proposed design and I was wondering if the
professionals on this board could set me straight as I am not an OLAP
expert.

The goals of the project must accomplish two things:
1. Overall trending and analysis of key performance
indicators. (Director and above type reports.)

2. Provided the lowest level of detail possible. We must
be able to view record level detail. For example a listing of all the
account numbers sold by a rep for any given time frame or a listing of
the hours worked by a rep day over day for any given time frame.
Basically I need the data before any aggregations are applied.
(Supervisor and manager level reporting.)

The proposed solution is to abandon an existing SQL Server database
for an OLAP only solution - Essebase. The project implementer says
that OLAP can provide the low level details if the cubes are set up
correctly.

Is this correct? I thought OLAP aggregated data into cubes thereby
making trending quicker. Six months of data is about 10 gigs in SQL
server. My fear is an MDBS design that tries to meet both objectives
would be ludicrous. Again my knowledge of OLAP is limited. Please
set me straight.



Thanks in advance.

Reply With Quote
  #2  
Old   
Nigel Pendse
 
Posts: n/a

Default Re: OLAP design question - 09-17-2004 , 02:30 AM






"csharp_c0d3r" <crystal (AT) captivatedminds (DOT) com> wrote

Quote:
To all OLAP gurus:

My organization is considering an OLAP solution for business
intelligence/reporting. I think this is great but I have serious
misgivings about the proposed design and I was wondering if the
professionals on this board could set me straight as I am not an OLAP
expert.

The goals of the project must accomplish two things:
1. Overall trending and analysis of key performance
indicators. (Director and above type reports.)

2. Provided the lowest level of detail possible. We must
be able to view record level detail. For example a listing of all the
account numbers sold by a rep for any given time frame or a listing of
the hours worked by a rep day over day for any given time frame.
Basically I need the data before any aggregations are applied.
(Supervisor and manager level reporting.)

The proposed solution is to abandon an existing SQL Server database
for an OLAP only solution - Essebase. The project implementer says
that OLAP can provide the low level details if the cubes are set up
correctly.

Is this correct? I thought OLAP aggregated data into cubes thereby
making trending quicker. Six months of data is about 10 gigs in SQL
server. My fear is an MDBS design that tries to meet both objectives
would be ludicrous. Again my knowledge of OLAP is limited. Please
set me straight.
You can drill to the detail stored in the RDBMS in most OLAP servers,
including Essbase. In other words, the cubes themselves don't need to
store the lowest level of detail, which you leave in SQL.

Of course, if you currently use SQL Server, you could use Analysis
Services which comes with it. This provides similar aggregation options
to Essbase, and will save you from having to buy/use another product.




Reply With Quote
  #3  
Old   
JM
 
Posts: n/a

Default Re: OLAP design question - 09-25-2004 , 02:15 PM



I agree with Nigel's suggestion. Setting up an OLAP reporting solution in
MSAS would probably be a good idea.. Essbase ain't cheap, and you are
introducing a new system, more integration points, Essbase can be a
difficult system to support, etc.


"Nigel Pendse" <nigelp.nospam (AT) compuserve (DOT) com> wrote

Quote:
"csharp_c0d3r" <crystal (AT) captivatedminds (DOT) com> wrote in message
news:2822e169.0409161522.51a4170c (AT) posting (DOT) google.com
To all OLAP gurus:

My organization is considering an OLAP solution for business
intelligence/reporting. I think this is great but I have serious
misgivings about the proposed design and I was wondering if the
professionals on this board could set me straight as I am not an OLAP
expert.

The goals of the project must accomplish two things:
1. Overall trending and analysis of key performance
indicators. (Director and above type reports.)

2. Provided the lowest level of detail possible. We must
be able to view record level detail. For example a listing of all the
account numbers sold by a rep for any given time frame or a listing of
the hours worked by a rep day over day for any given time frame.
Basically I need the data before any aggregations are applied.
(Supervisor and manager level reporting.)

The proposed solution is to abandon an existing SQL Server database
for an OLAP only solution - Essebase. The project implementer says
that OLAP can provide the low level details if the cubes are set up
correctly.

Is this correct? I thought OLAP aggregated data into cubes thereby
making trending quicker. Six months of data is about 10 gigs in SQL
server. My fear is an MDBS design that tries to meet both objectives
would be ludicrous. Again my knowledge of OLAP is limited. Please
set me straight.

You can drill to the detail stored in the RDBMS in most OLAP servers,
including Essbase. In other words, the cubes themselves don't need to
store the lowest level of detail, which you leave in SQL.

Of course, if you currently use SQL Server, you could use Analysis
Services which comes with it. This provides similar aggregation options
to Essbase, and will save you from having to buy/use another product.





Reply With Quote
Reply




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off



Powered by vBulletin Version 3.5.3
Copyright ©2000 - 2012, Jelsoft Enterprises Ltd.