Private? Public? Hybrid! Accelerating Cloud Adoption
Cloud adoption is accelerating dramatically. After several years of experimentation, enterprises are transitioning to a new phase of extended deployment. The first phase of experimentation, where many early adopter enterprises “played” with multiple public and private clouds, is over. Now the early majority is getting set for the cloud journey, and increasingly that involves hybrid cloud adoption.
According to Sajai Krishnan, VMware’s vice president, cloud management, it is a conclusion driven by the realization that the act of leveraging and migrating to the cloud is more complicated than companies first thought. “Enterprises have recognized that private cloud is hard, public cloud is not easy,” Krishnan says, “and the reality is that for most, the future is a hybrid of multiple clouds—public and private.” They are realizing that new cloud management models are needed to make this transition successfully.
“The pace of innovation is accelerating,” Krishnan says. “Enterprises are looking for new models that can leverage the pace of this innovation while trying to be more agile and innovate at greater speeds.”
Commoditizing Application Development
“This is a great time to be software engineers,” says Krishnan. “Far from investing time in building infrastructure and dealing with other development environment vagaries, engineers are using standardized APIs and a multitude of ready-to-use services to deliver solutions that directly benefit customers. And they’re able to do this by standing on the shoulders of giants—giant platforms, that is.” These cloud platforms allow developers to be more productive and deliver customer innovation at ever-increasing speeds while staying focused on higher order needs.
“With Amazon Web Services (AWS), and a number of other cloud leaders,” Krishnan says, “if the developer knows how to use that model, a whole set of that infrastructural complexity just disappears.” This allows the developer to ignore any infrastructure issues and focus innovation on the application itself.
Beyond traditional infrastructure availability of ready-to-consume services, software development itself has seen massive innovation. Krishnan explains that “micro-services, containerization including technologies such as Functions as a Service (FaaS), are getting to be mainstream and competing with canned application development platforms and frameworks.”
The result is a development model that commoditizes the application development process by making that process separate and independent from the underlying infrastructure. In short, it is a very dynamic and agile world when it comes to software development.
The Pace of Innovation
To understand how revolutionary this is, look no further than recent history. VMware originally made its name with virtual machines (VMs). Just a few years ago, containers were the rage and the Docker ecosystem appeared like it would become developers’ choice for future app development. But no sooner had Docker become popular than Kubernetes and its overlaying container orchestration framework became the popular consumption model. And now, FaaS and even newer application development models, characterized by examples such as Amazon Lambda and OpenWhisk, are becoming the new application development flavors of the month.
“The paradigms are changing, so what you need to have now is a foundational infrastructure layer that can support these new consumption models,” Krishnan explains. “Ultimately no one cares about the platform. It’s all about the software value you deliver to your customer.” The challenge is to allow the applications to run over any cloud platform. Driving home this point, Krishnan points out that banking customers couldn’t care less how they are able to scan a check to cash it.
“During phase one, the experimentation phase of cloud, development teams invested in figuring out various agile infrastructure platforms,” Krishnan says. “The reality has now sunk in, and it’s clear to management that scarce development resources must deliver customer application value and not be distracted by cool infrastructure platforms.” The challenge for businesses now is having the capability to work with this plethora of cloud platforms and not lose sight of the primary mission of their organization.
The Cloud Management Anchor
The reality is that all of these frameworks—cloud platforms, application development models—run on some server, storage medium, and network. The ideal cloud management solution would manage that underlying infrastructure in a multi-cloud context and in a flexible way, whether that infrastructure is on-premises, off-premises, or a hybrid of both.
VMware’s cloud management platform provides that flexible, or what Krishnan refers to as “low-friction,” approach by automating the cloud management processes. VMware’s approach allows organizations to automatically assess the health of their public and private infrastructures to ensure that developers have the capacity, performance, and security they need.
Next, the platform provides IT and development teams easy access to commonly consumed infrastructure and services necessary for rapid application development and deployment, while ensuring organizational requirements around access, compliance, and governance are met.
Lastly, the system also monitors the cost of these operations to ensure that workloads are directed to the most efficient private or public cloud location.
The Best of Both Worlds
A strong point of the VMware cloud management suite is that it allows organizations to define a large array of policies for the business, rather than on a case-by-case basis. Everything is automated and based on different rights and privileges, so it all moves smoothly. “When you go to multi-clouds today, you can get the agility you want,” Krishnan says, “but you lose the control.” VMware technology provides the best of both worlds: agility and fine control.
Krishnan says that the ultimate goal of VMware cloud management is to deliver a platform that is similar to the vision of self-driving vehicles. Applications and workloads would run independently over any cloud without the need for “hands-on” supervision. It would all be automated and allow organizations to focus on providing better customer service, rather than worrying about changing the tires on the (cloud management) car.
“We’re not yet in the self-driving vision,” Krishnan says, “but we’re getting there.” In 2012, few would have believed that the self-driving future was imminent, and self-driving cloud management is similarly progressing just as fast. “Many customer segments will quickly realize the underlying value of the VMware cloud management platform,” Krishnan concludes. “They won’t see VMware. That is the point. We’ll be there in the background.”