Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations SkipVought on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Cognos Pros Cons 1

Status
Not open for further replies.

esearing

IS-IT--Management
Aug 22, 2000
132
0
0
US
Evaluating Cognos and have seen their show. What are the realworld strengths vs weaknesses? Can one truly query a cube with 1Million records in a matter of seconds as Cognos would have me believe. Anyone using for large scale monthly Telecom reporting (or large banking or similar transactional based business)? Looking at PowerPlay, Impromptu, and QC.

data resources:
3 Oracle Data warehouses with 50+ Million Records/Transactions per month raw data. moderately complex data structure with 200+ tables.
1 data mart contains monthly summaries for quick processing, roughly 500K records per month. some tables relatively static.

Current environment Windows moving to HP Unix.


[ur]www.eSearing.com[/url]
version 3
Photography, Flash Experiments, Stuff.
 

I've done a number of large scale banking and insurance implementations, and PowerPlay has had similar "in seconds" performance to what your salesrep has indicated.

Your sales rep should have done a demo showing a pyramid of degree of summarization. PowerPlay works well for mid-to-high level analysis. Impromptu is the appropriate tool for transactional level reporting and analysis (these transactions for customer xyz)
 
I would prefer a NON VENDOR response. Vendors don't tend to point out weaknesses. [ur]www.eSearing.com[/url]
version 3
Photography, Flash Experiments, Stuff.
 
Once the cube is built, the performance is good, but building the cube can last a very long time (can be several hours). Also note that during the cube-built big temporary files are built (5 GB).
At a certain volume of data, we were no longer able to build a cube. In this case, additional tuning must be applied, but I have no experience with this.
Remark : we build our cubes on Unix (AIX)
 
Hi,

Whatever is being said above is correct. Cognos does provide data in seconds. It does takes long time to create cube and it also creates huge log files.
To work on cognos, an ideal database is the one which follows dimensional model. If your Database is dimensional then you dont need to worry about anyhting. I used to work in a bank where the database was not dimensional and we had a tough time building the cubes. It may take ur full working day to create the cube.

Vish

PS: Use unix as ur server
 
The speed of Impromtu also depends on the database connection. We tap into our ERP system using OBDC, as our system is in Progress. Using OBDC is very slow, comapred to direct connecting to the database.

Our Cubes take about 30-50 minutes to build (we don't have the volume of data esearing has), but when I exported the raw data to txt format & re-did the same cubes they ran in less than one minute.

Once the cube is built, it opens in milliseconds however.

Bruce
 
Hi,

On Cognos' strengths, the top one I would stress is security. They are easily the best in the industry (IMHO). Most other companies talk about security as in who can see what reports or cubes. Cognos products go to the level of allowing dynamic filtering of data-level access within reports and cubes, a much more powerful set of options.

Most of my product experience is with Impromptu rather than PowerPlay. Impromptu is an excellent tool for simple to complex reporting. Speed is largely a function of report design and database platform choices. There are no inherent performance obstacles in the Impromptu product itself. Occasionally I hear of complaints from some users about speed, but it usually relates to filters written so elements are excluded from the SQL call to the database by use of local functions (forcing Impromptu to apply the filter locally), or other cases where it relates to the ability of Impromptu to do outer-join processing locally when the database cannot do it, a possibility that few other reporting tools can begin to attempt.

My 2 cents. Get back with specific questions if you desire.

Regards,

Dave Griffin
The Decision Support Group
Reporting Consulting with Cognos BI Tools
"Magic with Data"
[pc2]
Want good answers? Read FAQ20-2863 first!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top