IBM Melds Data Management with Kubernetes
IBM this week made available a data management platform built on microservices deployed on top of Kubernetes.
Rob Thomas, general manager for IBM Analytics, says IBM Private Cloud (IPC) for Data now also supports MongoDB and Postgres databases in addition to providing previously promised support for Db2 Event Store in-memory database.
IPC for Data can run in both a public cloud or on-premises and is intended to enable IT organizations to modernize data management by taking advantage of containers to simplify deployment. IT also can be deployed on multiple distributions of Kubernetes, including the one employed by Red Hat in its OpenShift platform-as-a-service (PaaS) environment.
The IBM framework is designed to ingest and analyze data in real time using an event-driven architecture based on an instance of the Apache Spark in-memory computing framework and the Apache Parquet Data Format, which is used to facilitate building of AI models constructed using various types of machine and deep learning algorithms.
IBM also is making available an additional suite of complementary tools and services, including IBM Data Science Experience, Information Analyzer, Information Governance Catalogue, Data Stage, Db2 relational databases and Db2 Warehouse.
In addition, IBM has tightened the integration between IPC for Data and its risk management service. In effect, Thomas says organizations that embrace IPC for Data become immediately compliant with regulations such as the General Data Protection Rule (GDPR) that require organizations to be able to more granularly maintain control over their data.
Longer term, Thomas says that as IT organizations continue to take advantage of platforms such as IBM Data Catalog to provide a layer of governance on top of their data, it will be only a matter of time before DataOps becomes more tightly coupled with DevOps to reduce the amount of friction associated with building and maintaining pipelines within applications. As that process occurs, Thomas notes there inevitably will be a corresponding increase in the willingness organizations will have to build and test new applications that leverage their data assets.
It’s still early days it terms of marrying modern microservices-based platforms to data management. But as it becomes easier to deploy databases and associated data management tools on top of platforms such as Kubernetes, the opportunity to distribute those technologies across the enterprise increases exponentially. In effect, data management will need to become more federated. That capability, in turn, will make it easier to bring application code to where data resides while still being able to centrally manage all that data, says Thomas.
Over time, databases and applications servers that organizations employ today to manage all that data themselves will become less monolithic. Much like the applications built on top of them, databases and application servers soon will be deconstructed into a series of data stores based on microservices that will be eaier to distributed and invoke.
In the meantime, it’s clear Kubernetes and microservices are about to transform data management in ways that should prove over time to be much more DevOps friendly.
