Recommendation on TM1S.CFG values

Product:

Cognos TM1 10.2.2 Fix Pack 6

Microsoft Windows 2008 R2 server

Problem:

What is a recommended starting point for settings in the TM1S.CFG file?

Solution:

In the TM1S.CFG file that you use for new TM1 applications instances, the settings vary with the server capacity and the need of the application. Here is a suggestion:

 

MTQ=4

If you have Microsoft Windows server with 8 CPU cores, and on that server, you run other TM1 instances or the TM1 Application web, you should not utilize all cores. I would recommend to start with MTQ=4 and then adjust it to MTQ=6 if the load allows it.

Query caching behavior is configured per cube via the VMM value in the }CubeProperties cube, where the VMM value defines the maximum amount of memory to be used for caching per cube. If memory permits on server a new start value could be 512 kb.

If VMM cache is set too low, even queries that were cached without MTQ use may not be cached anymore once MTQ is enabled. In such cases – to avoid unnecessary re-execution of MTQs – increase the VMM value until repeated query execution will not trigger MTQ activity anymore (indicating the cache is used).

 

To enable multi-threaded start-up, a configuration parameter must be added to tm1s.cfg file: MaximumCubeLoadThreads=<#ofThreads>.

IBM recommends starting with 50% of the number of available processor cores for your server and testing the trade-off of additional memory consumption and start-up time to determine the optimal setting for this configuration parameter.

When MaximumCubeLoadThreads is set to 0, cube loading and feeder processing is NOT multi-threaded. This is the default behavior when MaximumCubeLoadThreads is not explicitly set in the Tm1s.cfg file.

If your Cognos TM1 model uses conditional feeders where the condition clause contains a fed value, you should set MaximumCubeLoadThreads=0 or exclude the parameter from the Tm1s.cfg file to disable the use of multiple threads at load time.

 

PersistentFeeders=T

To improve reload time of cubes with feeders, set the PersistentFeeders configuration parameter to true (T) to store the calculated feeders to a .feeders file. The feeders will be saved to a *.feeders file (per cube) when the data is saved or rules are edited. On Startup and with PersistentFeeders=T, TM1 will load feeders from the saved feeders file which reduces the time normally taken to recalculate those feeders.

 

ViewConsolidationOptimizationMethod=TREE

the tm1s.cfg parameter ViewConsolidationOptimizationMethod defines the method used to achieve view consolidation optimization. Recommended to have it to TREE.  For example, in the uncommon circumstance when dimensions have just a few leaf elements rolling up to many consolidations, ViewConsolidationOptimizationMethod should be set to ARRAY .

 

Recommend to have a ClientMessagePortNumber set to ensure good performance on TM1 Architect in Citrix environments.  If you are running more than one TM1 server on a machine, each must have a unique port number noted.

ClientMessagePortNumber=<your port number>

RuleStatusIncludeConditionalCheck=F , can improve performance in TM1 10.2.2

LockPagesInMemory= F (False to allow windows to handled memory)

In rare circumstances, should this be set to T. Recommendation is to set it at F.

If a Cognos TM1 server running on a Windows 64-bit operating system is idle for a long period of time, physical memory taken up by the Cognos TM1 server will page out to disk. This is a function of the Windows 64-bit operating system and not Cognos TM1 . This can cause performance degradation in large Cognos TM1 databases when trying to access data after an idle period.

To maximize performance when running a large Cognos TM1 database on 64-bit Windows, set LockPagesInMemory=T in the Tm1s.cfg file.

When this parameter is enabled, Windows still trims pages from the Cognos TM1 process space, but does not page them to disk. This benefits Cognos TM1 server performance because objects are no longer placed in virtual memory, but instead remain in physical RAM.

 

For Performance Modeler to work you need:

AllowSeparateNandCRules = T

DistributedPlanningOutputDir=tunit

ForceReevaluationOfFeedersForFedCellsOnDataChange=T

 

For SSO with CAM you need:

ServerCAMURI=http://biservername:9300/p2pd/servlet/dispatch

ClientCAMURI=http://biservername/ibmcognos/cgi-bin/cognosisapi.dll

ClientPingCAMPassport=900

Indicates the interval, in seconds, that a client should ping the CAM server to keep their passport alive. If an error occurs or the passport expires the user will be disconnected from the Cognos® TM1® server.

 

CAMPortalVariableFile=portal\variables_TM1.xml

CAMPortalVariableFile parameter needed only when variables_tm1.xml file name changes.

https://www.ibm.com/support/knowledgecenter/en/SS9RXT_10.2.2/com.ibm.swg.ba.cognos.tm1_inst.10.2.2.3.doc/c_camportalvariablefile_1.html#CAMPortalVariableFile_1

 

How to setup Cognos Access Manager authentication

https://allthingscognos.wordpress.com/2014/08/26/configuring-performance-modeller-and-tm1-web-with-cam-security-for-tm1-10-2-n/

 

Other information:

The IBM® Cognos® TM1® server may crash when Parallel Interaction is turned on and Bulk Load mode is used. In version 10.2.0 of Cognos TM1, Parallel Interaction is turned on by default.

http://www-01.ibm.com/support/docview.wss?uid=swg21646056

Text is copied from different guidelines by IBM

https://www.ibm.com/support/knowledgecenter/SS9RXT_10.2.0/com.ibm.swg.ba.cognos.tm1_inst.10.2.0.doc/c_parametersinthetm1s.cfgfile_n1503fe.html