Planning Analytics

 View Only
  • 1.  The Cloud Containerization of TM1 Server Webinar

    Posted Tue May 17, 2022 11:28 AM
    Edited by System Fri January 20, 2023 04:12 PM
    Summary

    The Cloud/containerized version of TM1 server is coming soon with the next release of Planning Analytics for Cloud Pak for Data. Stuart King, Senior Product Manager for Planning Analytics with Watson, will present an interactive session where he will discuss the benefits of a containerized architecture and how companies can prepare to take advantage of it. Some of the key topics to be covered are:
    • Overview of containerization
    • Modernize your TM1 model for the hybrid cloud world
    • Improve uptime of TM1 databases with OpenShift

    Please join us in this on-demand recording. Share your questions below and register to watch here

    ------------------------------
    Stuart King
    IBM Planning Analytics Offering Manager
    ------------------------------
    #PlanningAnalyticswithWatson


  • 2.  RE: The Cloud Containerization of TM1 Server Webinar

    Posted Wed May 18, 2022 04:30 PM
    Stuart,

    Thanks for doing this and I am looking forward to it.

    ------------------------------
    Walter Coffen
    Technology Manager
    QueBIT Consulting, LLC
    ------------------------------



  • 3.  RE: The Cloud Containerization of TM1 Server Webinar

    Posted Wed June 01, 2022 02:19 PM

    Hi everyone,

    You can watch the on-demand recording here and download the slides here.

    Please share any of your questions below.



    ------------------------------
    Stuart King
    IBM Planning Analytics Offering Manager
    ------------------------------



  • 4.  RE: The Cloud Containerization of TM1 Server Webinar

    IBM Champion
    Posted Thu June 02, 2022 01:10 AM
    Edited by System Fri January 20, 2023 04:21 PM
    Hi Stuart,
    Thanks again for putting together and delivering the session.  A few questions below come to mind. I know one or two may have been answered but it may be good to have the responses here for anyone who may have missed them.

    1. Will there be a "readiness" tool to assess items that need to be addressed before you can migrate e.g. TI with ExecuteCommand, unsupported MDX, references to control cubes that have been deprecated?
    2. How will promotion of Dev to to QA to Prod be handled - Not only TI but other objects e.g. Dims, Cubes, Rules etc. All via REST API, GitHub, Lifecycle Management, File upload tool and REST API?
    3. What MDX changes are being made and what will no longer be supported / work?
    4. Will Native cube views fall aways?
    5. With containers not being for everyone and typically an enterprise requirement, how would smaller clients migrate to PA Engine/Next Gen to take advantage of new features in the future?
    6. Email would need to be driven via the ExecuteHTTPRequest - Is IBM planning on creating a simple module that could be called from TI where you simply pass parameters for To, Subject, Body, Attachments, much the same way we do this now but leverage PowerShell to send the mails?
    7. At what point/when would SaaS customers be upgrade/migrated to PA Engine/Next Gen? There is obviously some preparation required and code changes and we would need to plan accordingly on timelines, costs etc.
    Thank you.

    ------------------------------
    George Tonkin
    ------------------------------



  • 5.  RE: The Cloud Containerization of TM1 Server Webinar

    Posted Thu June 02, 2022 11:39 AM

    @George Tonkin

    Will there be a "readiness" tool to assess items that need to be addressed before you can migrate e.g. TI with ExecuteCommand, unsupported MDX, references to control cubes that have been deprecated?

     The TM1 to Planning Analytics Engine Migration Utility will address some of this.  For example, names of objects in the model that are unsupported by ODATA will be identified.  I don't believe we are currently to example MDX saved in views and sets to check for references to removed object names.  We can  consider an enhancement to the migration utility to check for references in MDX, rule, and Turbo Integrator scripts to cubes (e.g }CubeProperties) that no longer exist.

     

    How will promotion of Dev to to QA to Prod be handled - Not only TI but other objects e.g. Dims, Cubes, Rules etc. All via REST API, GitHub, Lifecycle Management, File upload tool and REST API?

     The Git ODATA API endpoints will still exist in Planning Analytics Engine.  It's not yet determined how we will expose this capability in the Workspace Modeling Workbench.  The backup and restore feature of Planning Analytics Engine can also be used to migrate a whole model between environments.

     

    What MDX changes are being made and what will no longer be supported / work?

    MDX in Planning Analytics Engine is the same as MDX in TM1 v11.  We are slightly changing the way security works with MDX.

     


    Will Native cube views fall always?

    Native views still exist in Planning Analytics Engine.  No change yet.  There are a few features we need to complete before we consider removing native views (e.g drill thru to MDX and building MDX views with Turbo Integrator functions). 

     

    With containers not being for everyone and typically an enterprise requirement, how would smaller clients migrate to PA Engine/Next Gen to take advantage of new features in the future?

    Planning Analytics Engine will also (eventually) be available as part of a Planning Analytics SaaS offering.  

     

    Email would need to be driven via the ExecuteHTTPRequest - Is IBM planning on creating a simple module that could be called from TI where you simply pass parameters for To, Subject, Body, Attachments, much the same way we do this now but leverage PowerShell to send the mails?

     We will provide sample Turbo Integrator code to show how the ExecuteHTTPRequest feature can be used with services, including an e-mail service.  We are not planning on building an e-mail service into Planning Analytics Engine.  We would consider providing a preconfigured e-mail service as part of a SaaS offering with Planning Analytics Engine.

     

    At what point/when would SaaS customers be upgrade/migrated to PA Engine/Next Gen? There is obviously some preparation required and code changes and we would need to plan accordingly on timelines, costs etc.

    There are no plans to deprecate the current Planning Analytics SaaS offering or Planning Analytics Local.  It's probably a bit early to provide any timeline.  However, we can talk about the differences between TM1 and Planning Analytics Engine to help you understand the work that is required to migrate.



    ------------------------------
    Stuart King
    IBM Planning Analytics Offering Manager
    ------------------------------



  • 6.  RE: The Cloud Containerization of TM1 Server Webinar

    IBM Champion
    Posted Thu June 02, 2022 11:43 AM
    Thanks for all the feedback Stuart - very useful to know.

    ------------------------------
    George Tonkin
    ------------------------------



  • 7.  RE: The Cloud Containerization of TM1 Server Webinar

    Posted Fri June 03, 2022 01:52 PM
    HI -
    Reading the part below: 

    How will promotion of Dev to to QA to Prod be handled - Not only TI but other objects e.g. Dims, Cubes, Rules etc. All via REST API, GitHub, Lifecycle Management, File upload tool and REST API?

     The Git ODATA API endpoints will still exist in Planning Analytics Engine.  It's not yet determined how we will expose this capability in the Workspace Modeling Workbench.  The backup and restore feature of Planning Analytics Engine can also be used to migrate a whole model between environments.


    I need to ask for some additional clarity. 
    In the new version, we will still be able to promote individual objects (TI, Dims, Cubes), even if it is different than how we do it today, but the ability will still exist, correct ?  
    We sometimes have more than one project going on in DEV at a time, with different scope/duration/etc.  We would still need to be able to promote certain objects and not others. In fact, if others are promotes too soon, they may not work, etc.  There are legitimate business reasons to not promote the entire model every time. 
    Also - being able to promote the individual objects, test, and document the test results specifically for exactly what changed and when is a critical piece of our ASCA-approved checkpoints in the promotion process. 
    Thank you

    ------------------------------
    DONA PESCHKO
    ------------------------------