IBM z/OS V2.5 integrates mainframes with hybrid clouds, AI

IBM put another piece of its hybrid cloud initiative in place with a new version of its mainframe operating system that works across multiple clouds and speeds development of AI-based modernization projects.

Version 2.5 of the z/OS operating system has new features that make it easier to use IBM z/OS Container Extensions and give developers access to a wider choice of AI-based tools, frameworks and libraries that include Linux-based container applications they can deploy within the z/OS platform. Other new AI capabilities include support for tools such as TensorFlow and IBM Watson Machine Learning for z/OS.

With the looming threats posed by recent ransomware attacks, IBM also focused on strengthening cybersecurity capabilities by expanding its pervasive encryption to new types of data sets, such as sequential basic format and large format SMS-managed data sets. This allows users to encrypt data without the need to make any changes to applications as well as to more simply address compliance issues.

Some see the new version as another attempt by IBM to make the mainframe a more integral contributor to corporate users' hybrid cloud environments and less a technology outlier.

"Users want to move toward cloud-native environments and leverage containers and things like OpenShift," said Judith Hurwitz, president of Hurwitz & Associates LLC, a research and consulting firm focused on emerging technology. "The easier it is for users to incorporate hybrid cloud environments without having to leave their [mainframe-based] world, the better. It makes the mainframe a more viable platform for the long term."

The majority of existing users bought into mainframes for the scalability, security and management capabilities of the machine in the age before hybrid clouds and AI, when there was little concern for tying their machines to those technologies.

The easier it is for users to incorporate hybrid cloud environments without having to leave their [mainframe-based] world, the better. It makes the mainframe a more viable platform for the long term.

"Ten years ago, Z users weren't part of any other ecosystem. It was strictly a standalone system," Hurwitz said. "But adding OpenShift and doing more cloud-native things, [users] are beginning to see the possibility of the mainframe being used as another platform for managing workflows."

The new security capabilities tacked on to the mainframe's pervasive encryption scheme are related to the added vulnerabilities hybrid clouds typically present to businesses. Providing security across multiple cloud layers often slows hybrid cloud implementations.

Besides the sequential basic format and large SMS-managed data sets, IBM added anomaly mitigation features that make use of predictive failure analysis, runtime diagnostics and a workload manager to better detect anomalous behavior in real time so users can take preventative steps to avoid availability problems.

"One of the impediments to hybrid clouds is the multiple layers of security IT must deal with," said Frank Dzubeck, president of Communications Network Architects, Inc. "IBM Research has been focusing on bridging the gaps in all these different layers so users can layer regulatory policies across not just the cloud stack, but the on-prem stack and the edge stack. This [new version] is built to address this."

The new release also underscores IBM's stated goal of establishing a unified data and AI platform by tying together cloud, on-premises and edge platforms; reducing data silos, such as mainframes; and encouraging AI adoption. It also backs up the company's idea behind the data fabric, which addresses another IT security concern.

"The idea behind the data fabric is to take data that solves a specific problem and put it into a lake or data warehouse and leave it there," Hurwitz said. "This eliminates the need to move mission-critical data all over the place, increasing the risk of having it intercepted by hackers. Instead you can now process data where it resides and not move it to another platform."

IBM z/OS version 2.5 also includes features to help legacy users more easily modernize their mainframe environments by allowing tighter interoperability with mainframe and cloud-based tools. The company has added new interoperability between Java and Cobol by extending the two programming environments' support for parallel 31-bit and 64-bit, making modernization efforts easier to accomplish.

IBM also introduced new features to integrate cloud storage through transparent cloud tiering and support for object Access Method cloud tier support, which helps reduce operating expenses with data transfer to hybrid cloud storage environments.            

As Editor At Large with TechTarget's news group, Ed Scannell is responsible for writing and reporting breaking news, news analysis and features focused on technology issues and trends affecting corporate IT professionals.