Master Data Management

 View Only

Latest MDM release extends container orchestration, machine learning and operational capabilities.

By Marcus Boone posted Fri May 31, 2019 09:53 AM

Master data management continues to be a hot category, and its no wonder why.  Organizations of all sizes and maturity are on a journey to be more data and customer-centric in everything they do.  To be effective at this requires trusted information about your customers, patients and citizens and having the context necessary to serve them effectively in the era of the individual-centric economy.  That's why I'm especially excited to share some of the highlights in feature pack 9 of IBM InfoSphere Master Data Management 11.6 product line.

First up is our new applied machine learning for data stewards.  As good as our matching is, there is a risk of over or under matching records in any MDM system.  This means there is often a gray area that can only me reconciled by humans. But being too conservative means these tasks, or clericals, can back up into the thousands, making it a tedious if not daunting task to keep up with. But now, we are offering anew machine learning capability in Infosphere Master Data Management Standard Edition.  This capability provides assistance to the data stewards with their entity resolution decisions. It can essentially enhance the efficiency, consistency, accuracy and automation of suspect duplicate processing in your organization. This component indeed learns from the data stewards’ decisions. Its underlying model may be periodically retrained, in order to react to the new data patterns and to provide a smarter and more informed perspective. This capability is also offered as a Docker image, to support easy experimentation.

Secondly is the general availability of our Operational Cache.  More and more, successful organizations are tapping into their trusted master data as part of regular operations.  This means the reads on this system often account for 90% of the InfoSphere Master Data Management Advanced Edition workload.  It also means critical customer-facing applications are reliant on the availability of master data, making maintenance and scheduled downtime for your MDM system no longer an option.  Operational Cache addresses your organization's changing needs by providing very fast and scalable access to the master data for your mobile applications and other online channels through modern REST APIs. It continuously maintains the latest global or local master data, in a graph datastore on Cassandra within single or geographically-distributed data centers.

Third, as more and more organizations seek agility in their operations, they are looking to docker and private cloud infrastructures to support those initiatives.  On the heels of our December 2018 feature pack which introduced Kubernetes support for our docker containers for IBM MDM, we now support running MDM in Red Hat® OpenShift®environments.  Red Hat® OpenShift® is a comprehensive enterprise-grade application platform, built for containers with Kubernetes.  This brings a new deployment option to running IBM InfoSphere Master Data Management Standard and Advanced Editions.

Finally, Kafka has become a more prominent choice as organizations look to open source for messaging.  IBM InfoSphere Master Data Management Advanced Edition now publishes business events to Kafka. This allowfor more scalable, available and modernized integration between MDM and other integrated applications. For example, IBM DataSteward Center UI and IBM Business Process Manager (BPM) are now enhanced to acquire data steward tasks from Kafka. We have also built a new free-text search capability using Kafka and streaming and Elastic Search.

Interested to know more? View the knowledge center for a full list of what is new, or drop us a comment here.