Cloud Pak for Data Group

What's new in Cloud Pak for Data 2.1

By Shannon Rouiller posted Fri June 14, 2019 04:52 PM

IBM Cloud Pak for Data 2.1 screens

In addition to a new name, IBM Cloud Pak for Data (formerly IBM Cloud Private for Data) version 2.1 has exciting new features and enhancements.

Key enhancements

Multi-tenancy architecture
Different areas of your organization can share the same cluster, but can operate independently in a dedicated instance. Each instance has its own users, data, quotas, namespace, and ports.

Installation on AWS and Azure with Terraform
A Terraform installation bypasses a number of manual setup steps, so installing a stand-alone version of Cloud Pak for Data on Amazon Web Services or Microsoft Azure just got easier.

Easy rediscovery of enterprise data catalog assets
Previously, if you made changes to any discovered assets, you had to start from scratch to discover the changes and reflect them in the catalog. Now you can click the Discover again button to simply rerun the process with the same settings and update the catalog with the changes.
Runtime parameters for data transformation jobs
With runtime parameters in your data transformation jobs, now you have the flexibility to specify parameter values whenever you run a job instead of hard-coding an update into a job every time you want to change a value. 

New database features
  • Provisioning a Db2 Warehouse is now a lot faster.
  • New backup and restore capabilities are available for MongoDB and Db2 Warehouse databases.
New and improved add-ons

New IBM DataStage premium add-on
With the new IBM DataStage premium add-on, you can easily integrate, transform, and deliver data in batch and real time. With the add-on, you can expand your data transformation capabilities with:
  • Additional stages, such as hierarchical, column import,  column export, column generator, external source, and external target.
  • Additional connectors, including Apache HBase, Snowflake, and Salesforce.
  • Productivity features, such as job scheduling and job sequencing.
    New third-party add-ons 
    New add-ons are available from Knowis, NetApp, PostgreSQL, and WAND

    IBM Streams included and improved
    IBM Streams is now included in the Enterprise Edition and Cloud Native Edition offerings, so you no longer need to purchase the add-on to use it for processing real-time streaming data, including IoT data. In addition, new notebook samples are provided for integration with SQL databases, IBM Event Streams, and IBM Db2 Event Store.  Developers can also optionally provision a new lightweight instance of IBM Streams with an embedded version of ZooKeeper. 

    Enhancements to the Watson OpenScale add-on

    Watson OpenScale now has bias detection to automatically detect certain protected attributes in a model, accuracy metrics for the de-biased models, and support for the Microsoft Azure Machine Learning Studio framework.
    Data virtualization enhancements 
    Now you can virtualize data that resides in Excel and CSV files on remote file systems. Other new Data Virtualization features include query pushdown optimizations for Netezza, Hive, and Teradata data sources, and an optimized process model for ingesting large amounts of data.

    Learn more
    To read more about what's new in 2.1: Want more? Sign up for the 2.1 webinar "IBM Cloud Pak for Data v2.1: Innovation Continues with Exciting New Features," which will cover the new features for Cloud Pak for Data v2.1 as well as introduce the brand new IBM Cloud Pak for Data System appliance.