Cognos Planning Cubes Not Opening? It’s Your Memory!

4/10/15

​Cognos Planning Cubes Not Opening? It’s Your Memory!

As a Business Analytics Coach with over 15 years of experience implementing Enterprise Cognos Planning, I often get calls from clients saying, Cognos Planning Cubes Not Opening is a huge problem, please help!

IBM Cognos Planning has performance and memory issues when the size of the cubes or contributor models you’re opening includes large amounts of data. Cognos Planning was built on a distributed architecture. This means when the end users, contributors, open their models, it downloads the data appropriate to them to their machine. When the end user changes, the data system uses the memory of the end users machine to calculate the new values. The values are then sent back to the server when the contributor hits “save” or “submit”.

Cognos Planning Cubes Not Opening

Back in the day, the distributed architecture was how Adaytum, the predecessor to Cognos Planning, was a leader in deploying budgeting and forecasting solutions to thousands of end users over the web. But today, we want more data so we add items and dimensions to our models causing memory errors. Unfortunately, IBM Cognos Planning has size limitations. If you are trying to open multiple nodes at the same time, you are very likely to have memory issues.

Cognos Planning Cubes Not Opening - quick fix

Short term fixes include:

  1. Add more memory to the power user and end user machines
  2. Leverage the “No Data” setting on the access tables to cut down what is passed to the contributors. See IBM site for information on No-Data Setting - http://www-01.ibm.com/support/knowledgecenter/SSWGNW_8.0.0/com.ibm.swg.im.cognos.contrib_admin.8.4.0.doc/contrib_admin_id6035RulesforAccessTables.html%23RulesforAccessTables
  3. Add dummy nodes in your elist so you can open fewer nodes at one time.

This is where reality sets in. With today’s increased demand for data, the issue ends up being bigger than just adding more memory.

IBM Cognos Planning is outdated technology!

The good news is that IBM does have a solutions for you. TM1, also called IBM Performance Management, was purchased in 2007 for its tremendous calculation engine. Over the years, IBM has merged the things we love about IBM Cognos Planning and the powerful 64 bit in memory calculation engine of TM1. The result is a powerful solution that can be deployed to a large number of users, while eliminating the memory issues Cognos Planning has inherently. The calculations are done on the server…not the end users machine. TM1 allows for large cubes and faster response times. Lodestar Solutions has migrated many clients from Cognos Planning to TM1. Many of them were able to merge multiple Planning applications, which were broken down because of size issue, into a single TM1 model.

Benefits of TM1

There are so many benefits to moving to TM1 I can’t cover them all here. Lodestar Solutions created a series of videos on the benefits and how to migrate from Cognos Planning to TM1. To get access to our private library, go to www.MovetoTM1.com.

Link to Express demo: http://www-01.ibm.com/software/analytics/cognos/express/demos.html

If you would like to talk to one of our Cognos Planning/TM1 coaches about your specific situation, email us at Sales@LodestarSolutions.com or call 813-254-2040.

Moving from IBM Cognos Planning to TM1 Part 2

This is the second piece to the story on why an IBM client would consider moving from Cognos Planning to TM1. If you missed Part 1 of Cognos Planning to TM1,  you can read it by clicking HERE.

Power User Benefits in Moving from Cognos Planning to TM1

The power user license is the TM1 Modeler.  The role is similar the Enterprise Planning Modeler in that this person typically is responsible for building, modifying and deploying the TM1 Planning model to the end users (Contributors).

  • NO SYNC, GTP & RECONCILIATION-  In IBM Cognos Planning, changes are made in Analyst then the models are synced with Contributor and the Go To Production (GTP) process is run which then triggers a reconciliation.  Depending on the model and the job servers, this can take some time.  TM1 changes are available immediately to the end users, no sync and GTP required!
  • MODEL SIZE- Because TM1’s architecture is built on a 64 bit in memory, server based cube structure, TM1 can handle huge cube sizes.  Many of you may have broken down your Planning models into multiple applications over the years, well now you can put them back together again.  End users will be able to see all the data in one place without running links.   A TM1 cube can have up to 256 dimensions in one cube.  Try doing that in Planning!
  • NO LINKS – TM1 doesn’t need links because the rules allow the calculations to reference data in another cube without physically moving the data to another cube.  This resembles Excel where formulas allow calculations to use information in another worksheet.  This eliminates all the link maintenance and troubleshooting.
  • CALCULATIONS AT THE CUBE LEVEL –   In IBM Cognos Planning, calculations are really a combination of the formula at the d-list level and the links that feed the data into the calculation.  When you trouble shoot a situation, you are exploring both the calculation and the links.  However, in TM1 the rules are defined at the cube level making it easier to troubleshoot.
  • EASY TO DEFINE ONE OFF CALCULATIONS – Periodically, when you build models, you need a calculation that is unique by version or data point like employee type.  We can do this in Planning but requires some interesting calculations that often times require IID references.  In TM1 since the calculations are at the cube level, one off calculations are easy to define.
  • LESS OBJECTS TO MAINTAIN – TM1 models typically have fewer objects to manage such as the number of dimensions and cubes.  Links and a-tables are not needed.
  • DRILL THROUGH – TM1 Drills can be defined by the power users and deployed to the end users
  • SANDBOXES/PERSONAL SCENARIOS.  I love this feature.   End users can create their own versions on the fly.  This eliminates the need for them to call the admin requesting another version and waiting for a sync, GTP and reconciliation.
  • CONDITIONALLY FORMATED DROP DOWNS- If the model requires users to select from a drop down list, (similar to Planning dlist formatted items) the list can be conditionally formatted so they only see accounts that are applicable to their department or applicable to the other data they have entered.
  • NO PUBLISHING DATA FOR BI- Are you tired of waiting for the publish process to finish in order to update your BI reports?  TM1 provides real-time reporting with IBM Cognos BI.  Publishing is not required!
  • ALIASES – Power users can use aliases as another name for a data point. This can be very handy if you are pulling data from multiple GL’s with different account numbers.  You could have an alias for each GL.
  • NO NEED FOR HIDDEN ITEMS FOR CALCULATIONS – As stated previously, TM1 rules allow for calculations to reference cells in other cubes eliminating a lot of hidden items we have in planning.
  • WEIGHTED AVERAGES – TM1 allows for separate Consolidation and Details Rules making weighted averages a breeze.
  • ROLLUPS CAN HAVE MULTIPLE  HIERARCHIES- TM1 allows children to roll to different parents within the same dimension to provide multiple views of the same data (unlike Planning there is no need for another application to view data differently)

IBM continues to add improvements to the Power User’s experience with TM1.  As I described, there are a lot of Power User benefitsin moving from Cognos Planning to TM1.  In 2012, IBM is scheduled to release TM1 10.0.  All I can say is that you will LOVE IT! For more information on TM1 or any of the IBM Cognos modules, feel free to call us or email us at Sales@LodestarSolutions.com.

X