The cloud’s silver lining: the mainframe

How mainframes empower the hybrid cloud's success


Man and woman working with IBM type 704 electronic data processing machine used for making computations for aeronautical research, 1957.

Credit: Langley Research Center of the United States National Aeronautics and Space Administration (NASA)

In case you hadn’t noticed, 2016 has started with big momentum for mainframes – including IBM introducing the new z13s. It is always gratifying when something you believe in, against herd mentality, comes to pass. It is far too soon to say that skepticism about mainframes has passed, but the media and analysts are certainly telling a different story than they did a year ago. A major reason is a development I have been excited about for some time: the hybrid cloud.

It is no coincidence that positive press for the mainframe and the hybrid cloud are happening together. Here are just a few of the reasons why mainframes are at the heart of the mix that makes this hybrid so powerful.

Companies that rushed to embrace cloud computing during its hyped-up introduction have learned some hard lessons about its limitations, especially around security. The current move towards private clouds mitigates some of these risks, but when it comes to cybersecurity, “some” is never an acceptable answer.

The mainframe has always offered the inherent security of an onsite, single device, but new developments are taking advantage of Big Iron’s unique capacity to push data protection to new levels. Developers are harnessing mainframe prowess to process massive amounts of data at some of the highest levels of encryption available, at speeds other systems can’t handle. What other system can have a co-processor dedicated to cryptography? Add these expanding security abilities, including multi-factor authentication, to a cloud-integrated system, and you get a potent mix.

Bigger is better with a hybrid

A cloud system based on a set of linked servers can theoretically handle diverse types of computing tasks, as well as demands of scale. If only this was 100% true. The more discrete devices that are involved in a network, the more opportunities exist for malfunction. Today’s computing demands require the ability to process gigantic amount of data, at tremendous speeds, with virtually no room for error. Cloud systems alone may be able to tackle these demands adequately, but at what cost?

The marriage of mainframes and cloud networks in hybrid systems empowers the scalability of the cloud while providing a rock solid foundation for performance. The multiple processing structure of mainframes is perfectly suited for the varying bandwidth needs of tasks such as mobile transactions, and real-time analytics, to name just two. What better way to ensure speed than to have data and applications on the same device, while that same device is able to devote resources to each independently?

+ ALSO ON NETWORK WORLD When mobile and mainframes mix, the results are spectacular +

One of the major benefits of the cloud is storage, for cost, efficiency, flexibility. But when dealing with bigger than big data, other factors come into play. At a certain point, virtual storage does become less efficient than physical storage, because of the cost of transmission and the capacity for error from the network. This scenario, where the mix of cloud, physical backup medium, and mainframe is greater than the sum of its parts. New devices are emerging that allow mainframes to utilize the cloud for storage, but help overcome some of the cloud’s limitations around speed of backup and retrieval and extremely high thresholds for data integrity. Hybrid cloud storage further facilitates mainframe-based clouds by providing another option for the formidable task of mainframe data archiving.

A hybrid open to possibilities

The mainframe-based hybrid cloud is even more attractive given the explosion of open source as an element of mainframe design. Recent innovation in hardware and software design has made it possible to run an incredible array of operating systems and applications on mainframes, without compromising performance or security. Previously, cloud networks have offered developers a variety of unpleasant surprises around platform compatibility and throughput – simply because a network made of many machines also introduces many opportunities for error.

The opening of the mainframe also helps alleviate another major pain point of any IT department – making the diverse systems of a company, legacy and new, work together smoothly. A single compatibility glitch with a small but necessary application can impact the efficiency of the total network, no matter how it is distributed. A mainframe’s ability to dedicate processors to distinct operations, combined with its deep capacity to negotiate diverse platforms and applications, makes it a great solution for many a networking headache.

It would be foolish not to acknowledge that all of the above comes with a serious price tag. My response is that most advantages that mainframes offer to the hybrid cloud are features that have been part of Big Iron for decades. The long view is not so popular these days in tech, but I have to wonder how much money could have been saved by companies if, rather than spending on early versions of the cloud, they had invested in an expensive yet proven solution, like a mainframe?

This article is published as part of the IDG Contributor Network. Want to Join?

Must read: Hidden Cause of Slow Internet and how to fix it
View Comments
Join the discussion
Be the first to comment on this article. Our Commenting Policies