Db2 On Premise and Cloud

 View Only
  • 1.  DB2 Strategy for Analytics : BLU or WareHouse

    IBM Champion
    Posted Wed January 12, 2022 01:20 PM
    Hi Everyone,

    We are currently running DB2 11.1 for LUW and we want to modernise our analytics platform. The trade-up to Cloud Pak for Data seems the right choice, but there are some considerations that I think are not well explained. Should customer use BLU with their current db2 setup or should they consider to use Db2 Warehouse ? (this is for analytics)

    Is there any documentation on how to migrate from DB2 LUW to Db2 Warehouse (on-premises) ?

    Is there also any documentation from migrating from DB2 LUW to Cloud Pak for Data ?

    Thank you

    Largou walid

  • 2.  RE: DB2 Strategy for Analytics : BLU or WareHouse

    Posted Thu January 13, 2022 03:57 AM
    Hi Largu,

    the switch from Db2 (on prem) to Db2 Warehouse (on prem) is more a licensing issue and re-configuration issue, than migrating data as there is a Db2 underneath. It would be just backup/restore.
    But -- well -- for a performant DWH, you have to adapt the different architecture and partitioning, dockerization and so on. For Db2 11.1 there is still a description in the docs https://www.ibm.com/docs/en/db2/11.1?topic=analytics-migrating-data-warehousing-in-db2-version-111x which was not copied over to Db2 11.5 docs as the packaging of Db2, Db2 Warehouse and IBM strategy about CloudPak for Data changed. https://www.ibm.com/docs/en/cloud-paks/cp-data/4.0

    For moving to containers you could look at the new feature Click-to-containerize, found here https://www.ibm.com/docs/en/db2/11.5?topic=deployments-db2-click-containerize


    Roland Schock
    Distinguished Engineer
    ARS Computer und Consulting GmbH

  • 3.  RE: DB2 Strategy for Analytics : BLU or WareHouse

    IBM Champion
    Posted Thu January 13, 2022 06:51 AM

    Thank you for quick reply, for licensing we are currently targeting trade-ups for our customer to Cloud Pak for Data licensing from their traditional db2 deployments. My question is that for current users of db2 using it as their data warehouse, Is the recommended approach to migrate to Db2 Warehouse on containers or Db2 BLU on containers ?
    In each case, I think that the partitions are a must, as terabytes/petabytes databases cannot be containerised as single unit, going with multi partition MPP architecture is the right way.

    Thank you

    Largou walid

  • 4.  RE: DB2 Strategy for Analytics : BLU or WareHouse

    Posted Thu January 13, 2022 12:49 PM

    When we talk about containerization , we talk about the software and configuration.
    Most times a Db2 Warehouse is based on BLU technology.

    Partitioning overall with some considerations
    The decision to use a single instance or a partitioned environment is for me based on numbers of rows in a single relevant table. For the standard warehouses i would prefer single instance environment (rough estimate up to 10 TB).

    In an MPP environment you also have to determine a partitioning key to reduce the communication between partitions for joins. There
    When you partition you should also consider how much resources in CPU and Memory you want to spend for each partition. Maybe at minimum i would distribute over 4 partitions.
    These are only some points you have to think about. For some points tooling is in place or available to help.
    Some points can be covered throwing in resources like CPU and Memory.

    So maybe some discussions with more detailed information are needed.

    Kind regards

    joachim stumpf

  • 5.  RE: DB2 Strategy for Analytics : BLU or WareHouse

    Posted Fri January 14, 2022 09:12 AM
      |   view attached
    Hi Largou,

    About db LUW to db2 dtw (DPF) I can help you,
    the biggest private bank in Venezuela is our client since 9.7 DPF, DPF and in three months we are going to DPF on AIX.
    I hope this pdf will be useful for you.!

    Nicolas Rafael Ascanio Pena Data Consultant: Generalist, IBM Expert Labs
    Consultant Data & AI
    +58 424 1406472