Mainframes have been with us for over 50 years, but what we do with them and what we expect them to be able to do has been changing all that time. I thought it would be interesting to take a look at the next couple of trends that will be coming to a mainframe near you in the near future.
Once, all computing took place on a mainframe. There were remote devices connected to it, but all the number-crunching took place on a central computer—and you knew exactly where that was. But then Windows and UNIX machines sprung up and these little islands of computing worked on their own. Then everything became networked—a lot of it wirelessly. People used phones and tablets that had remarkable amounts of computing power. And these and laptops were all hooked to services that existed in the cloud. Sometimes, behind these cloud services were mainframes, and sometimes there were other servers. Then the Internet of Things (IoT) appeared, and there were even more devices that were networked to mainframes and other servers. And now, people are talking about edge computing.
So, what is edge computing? If we go back to our original model, all the computing took place at the center of the network on the mainframe. Even with the modern model, those IoT devices are reporting what they are sensing back to the cloud or mainframe, and programs running on the central servers decide what needs to happen next. It might be that the IoT device is monitoring the temperature of a greenhouse. The program decides that the temperature is getting too hot and it sends a message back to a device that can open the greenhouse’s window. Now all that to-ing and fro-ing of data takes time. So why not put some computing power at the furthest reaches of the network—on the very edge of the network and let that monitor the temperature of the greenhouse and control when the window opens and closes?
The benefits of doing this are that it allows analytics and knowledge generation to take place at the source of the data. Using edge computing significantly decrease the amount of traffic on the network. This can remove bottlenecks that may otherwise have occurred. Also, it can save money because there is less network traffic, and it’s much faster. Edge computing can also allow data to be pre-processed before it’s sent to the cloud or central servers.
Having thought about changing things at the edge of the logical network, the next change is at the very center—the heart—of the computing environment. This second trend is digital transformation. A recent whitepaper, The Business Value of the Connected Mainframe for Digital Transformation
, concluded that mainframe computing is at a crossroads—it will either continue supporting enterprise operations or it can play an increasingly important role in enterprise digital transformations. Executives from nine organizations described as running “significant mainframe operations” were surveyed to find out how their companies were making the most of their mainframes as part of their digital transformation initiatives, and also to see how valuable these efforts were. The modernized, “connected mainframe” integrates into an organization’s ecosystem—internally and externally—delivering innovations that drive revenue growth and improve operational efficiency, according to the whitepaper.
Mainframers will enjoy reading the whitepaper when it says: “IDC believes that the mainframe has a central role in digital transformation; businesses that do not take advantage of its broad range of capabilities are giving up value and, potentially, competitive advantage.” The whitepaper explains that: “Modernizing on the mainframe” is about “creating a platform that is integration-ready within the data center and with the outside world.”
The executives reported that as part of integrating mainframes into their digital transformation plans they made use of Linux on the mainframe to reduce costs, and used the cloud to enable mainframe as a service. They were also using Java. And they were using web services and service oriented architecture to communicate from the mainframe to other parts of the infrastructure and to deliver data and functionality within the mainframe to other applications. Respondents commented that, by adopting Java, internal APIs or Linux, they could “deliver innovation at a pace comparable to the rest of the IT organization.”
In addition, they said, to be successful, the mainframe must be connected to the rest of the data center infrastructure and IT processes and the outside world. There’s a need to expose services and capabilities on the mainframe to mobile apps. Many sites were making use of internal and external APIs on the mainframe. Other popular things to do were to use DevOps and Agile development on the mainframe. And many sites were starting to utilize hybrid cloud strategies.
On the downside, the report found that evolution of the mainframe is inhibited primarily by slow-to-change “cultural notions about the platform.” Many of the survey respondents expressed frustration with outdated perceptions in the organization that aren’t considering future potential. No surprises there.
So, looking at the two new ideas, we find that edge computing provides a way to get processing power out to the edges of the network; and digital transformation is all about staying up-to-date with the use of the mainframe and making it central to IT in an organization—and central to the thinking of executives who don’t currently understand it’s value or capabilities.
Trevor Eddolls is CEO at iTech-Ed Ltd, an IT consultancy. A popular speaker and blogger, he currently chairs the Virtual IMS and Virtual CICS user groups. He’s editorial director for the Arcati Mainframe Yearbook, and was an IBM Champion between 2009 and 2016.