Cloud Pak for Data

Is it time to change the conversation from Storage Services to Data Services?

  • 1.  Is it time to change the conversation from Storage Services to Data Services?

    Posted Wed October 20, 2021 05:32 PM
    Recently I posted a blog about what's new in IBM Cloud Pak for Data System 2.0.1 software that became Generally Available on October 12th. One of the topic I touched on in this blog is how Red Hat OpenShift Data Foundation Advanced in IBM Storage Suite for IBM Cloud Paks Starter Edition can help enterprise users to unlock the power within their data in new and impactful ways using data services. I want to elaborate on this a little more and open it up for discussion and comments from this community.

    Today, there are many storage services that we take for granted, for example, unlimited snapshots, clones, data replication and storage device performance monitoring and analysis. Most of these services were born when block was the primary access method, and they were carried over to files when File Systems became popular. These services were not designed to look inside the data (the content). They were created to simplify the work of system administrators to improve storage "container" management and all the intelligence to work with the content inside the data was left to the applications.

    Fast forward to the present day when object storage has become the primary data access method and unlimited number of attributes about the data can also be stored with the data. Scanning the content and analyzing it while a data object is being stored or accessed is the norm rather than the exception. Any operation on the data object can create an event or a trigger that can be passed on to the application. This can help unlock the power in the stored content in a number of ways, for example, for the purpose of auditing, compliance, access control, encryption, simplifying search, analytics and machine learning.

    I truly believe this is the biggest benefit of an object based storage infrastructure like Red Hat OpenShift Data Foundation, how it can simplify workload processing without the complexity of orchestration tool, hypervisor, container, operating system and so on. Data Services in OpenShift Data Foundation can help provide a plethora of workload specific services for data at rest, data in motion, data in action or cloud-native infrastructure services to any stateful application.

    I have spent a good part of my career in data storage. I have seen storage devices grow from 80MB to 16TB in capacity, and shrink from a stack of 14-inch magnetic disc platters to 3.5-inch solid state canisters in packaging, all at the same time. I often feel that innovation in data storage is just beginning, and we need to fasten your seat belts.

    What are you all seeing in the market and your organization, and what are your thoughts on the above?