Mainframes and cloud computing: back to the future

Cloud computing is one of the driving forces shaping many of IT’s most important initiatives. IDC has characterized cloud services as one of the pillars of computing’s next dominant platform, along with mobility, social media and big data.

As IT decision-makers evaluate next-generation solutions with an eye towards agility, security and cost efficiency, the cloud is inevitably part of the conversation: Should we use a public cloud service, a private cloud, a hybrid, or should we use a cloud infrastructure at all?

One of the most important questions that should be asked often is not asked: What kind of hardware platform should we use for our cloud solutions? Even at many of the largest cloud deployments, it is generally assumed that the cloud will be built upon a distributed infrastructure using industry-standard servers, typically X.86-based.

In reality, however, many cloud initiatives would be far better served using a technology that, in fact, has delivered all of the important functions of the cloud since well before the cloud was even a glimmer in anyone’s eye. That technology, perhaps surprisingly, is the mainframe.

If you think about it, the mainframe was the cloud before the “cloud” terminology was conceived. The mainframe has always handled mixed and diverse workloads concurrently, it has always offered elastic scalability and it has always been managed from a centralized platform. Remember: Virtualization was invented on the mainframe.

In fact, not only does the mainframe offer all of the critical functions required by cloud initiatives, it goes even further: The mainframe offers a more secure platform than any distributed platform, and the difference is fairly dramatic. Did you know, for example, that the IBM System z is the only platform that has never been breached on the Internet?

So why don’t more organizations utilize the mainframe for the cloud initiatives? There are a few ways to answer that question.

1. Some of the more savvy organizations actually are using mainframes for the cloud.
2. Recognizing how the mainframe can help deliver cloud efficiencies requires out-of-the-box thinking and overcoming pre-conceived (and misguided) ideas about costs and openness. In many cases, a mainframe will be far less expensive to deploy for the cloud than a distributed solution, and it will be far simpler to scale. It can also be built in a fully open Linux environment, requiring zero System z skills
3. Perhaps the best answer is this: As clouds get bigger and as scalability and security issues become more widespread and obvious, we expect to see an increase in organizations using the mainframe to support their cloud infrastructures.

At Vicom Infinity we have the knowledge and expertise to help our customers build any type of cloud solution – public, private or hybrid – on any IBM platform. We see tremendous value in exploring the mainframe as an alternative to traditional distributed platforms, and we can take you through a detailed and comprehensive cost analysis that will help you decide which platform makes sense for the type of cloud you need to build to address the specific needs of your organization.