The Hybrid Cloud Forecast: Part 2

In my previous post, I described how I got started with a podcast series called The Hybrid Cloud Forecast. In each episode, I invite an IBM Fellow as the guest and ask them to share their insights on hybrid cloud and their careers and experiences. Across the 30+ episodes published so far, three themes have emerged that appeared to come up:

• Research, which I covered in the previous blog
• Journey to Cloud, which I will cover here
• Life Stories, which I will leave for the next blog post

Cloud computing has impacted the IT industry like few other technologies. It has spawned an entire new industry of companies providing hardware and software for the cloud. Moreover, it has changed the way applications are built and run, and made additional, groundbreaking technologies possible and accessible (e.g., AI-based services).

No wonder, then, that it became the title of my podcast series—Hybrid Cloud Forecast—which is almost redundant since every aspect of cloud computing is hybrid in nature.

It isn’t a surprise that so many of the guests on my podcast work on topics and technologies directly related to cloud. One of the few pre-scripted questions I ask in most of the episodes is about the guest’s definition of “hybrid cloud.” The answers have all been comparable. Hybrid cloud is about:

• The location: In a data center, in a public cloud provider, on the edge
• The type of computing power: Commodity servers, highly specialized hardware like mainframes, GPU, quantum
• The set of architectural principles: Microservices, multitenancy, stateless-ness, serverless-ness
• The way applications are developed: DevOps, DevSecOps, many different programming languages, APIs
• The specialized clouds geared towards a certain industry: Government, financial services, etc.

In short, hybrid cloud impacts every aspect of where and how we run IT solutions.

Many of the conversations I had were about how exactly enterprises make that move towards hybrid cloud. It starts with the realization that virtually no one has a monolithic environment; everything is hybrid, and that must underpin the overall strategy.

• Vincent Hsu spoke about the revival of tape drives, and that a lot of innovation is currently directly towards that medium. Why? Because of the environmental impact of other media that cause much higher energy consumption.
• Rob High explained the increasing importance of running applications in non-traditional places and on non-traditional devices, which we also often call “on the edge.” Doing so comes with a need for new types of management tools and approaches, and it requires a greater degree of autonomy by those environments.
• Kyle Brown is the only guest who I had on the podcast twice. In the first episode (and it was actually the very first episode of the entire series) he talked about the evolution of different styles of application development and how cloud computing shaped that. And in his second appearance, he turned the table, switching from producer to consumer of application development solutions, triggered by his move from IBM Software to IBM’s CIO organization.
• Marcel Mitran, Rosalind Radcliffe and Kevin Stoodley (not published yet) all explained to me how relevant the mainframe is to cloud computing, especially in today’s hybrid world. Despite being in existence for several decades, mainframe technology is enjoying lots of innovation at a rapid pace.
• Tamar Eilam talked about how she came to the realization that what made cloud the disrupting force that it is was not the fact that it let you share computing resources outside your own data center; it was the automation that came with it that made it special.
• The only guest I had who is not an IBM Fellow is Naeem Altaf. When I saw his job title (“IBM CTO for Space Tech”), I thought that warranted an exception. He explained how running software in space—for example, in the International Space Station, in a satellite or in a rocket ship—is not very different from running software elsewhere in the cloud. The form factor can be considered a special case of edge computing, but the software is developed no differently. Microservices, containers, separation of state and logic—all of these architectural principles apply.
• Finally, Andrew Hately and Ajay Apte both talked about the challenges towards reduced carbon footprint and energy usage that need to be balanced against the seemingly ever-growing need for capacity in modern clouds.

Please see the Hybrid Cloud Forecast landing page to check out all the episodes we have to offer, and keep an eye out for the other two blog posts in this three-part series.

Read: “The Hybrid Cloud Forecast: Part 1“