Software-Defined Infrastructure Increases Agility

The rise of software-defined infrastructure may be one of those seminal changes to the ways enterprise technology is managed that comes along once in a decade. Unfortunately, it may take most IT organizations about half that long to implement SDI.

At its core, SDI seeks to eliminate the need to manually manage servers, storage and networks by making use of application programming interfaces (APIs) that make the entire IT infrastructure—deployed both inside and out of the enterprise—programmable. Once an SDI framework is in place, it enables the IT organization to respond faster to the changing needs of the business and, even more important, it fundamentally changes the way IT is managed.

Instead of managing individual silos of technology, SDI provides the opportunity to manage IT infrastructure at a higher level of abstraction. That capability, for example, makes it much simpler to dynamically manage all the servers, storage and networking resources associated with any business process in a much more holistic manner.

“It’s clear that IT infrastructure today is not optimized to meet the needs of the business,” says Krista Macomber, a senior analyst with Technology Business Research (TBR). “Software-defined architectures are not an endpoint in themselves. They are part of an ongoing effort to meet the needs of the business.”

In the first iteration, many of the SDI technologies that IT organizations had to master required the tech staff to learn how to program. That takes time and effort, and many IT professionals may not have the aptitude for writing code.

Embracing Converged and Hyperconverged Systems

To simplify that transition, Macomber reports that many IT organizations are embracing converged and hyperconverged systems, in which much of the required SDI technology is built into the system. Instead of learning how to program, IT staffs can use templates to program the environment without having to write code themselves.

“These systems are on ramps to a software-defined data center [SDDC],” Macomber adds. “Right now, the burden associated with managing infrastructure is too oppressive.”

Because of those requirements TBR is forecasting that hyperconverged platforms alone will represent a $7 billion market by 2020. Overall, Markets and Markets just published a research report that forecasts the SDDC market will grow from $25.61 billion in 2016 to $83.21 billion by 2021, representing a compound annual growth rate of 26.57% from 2016 to 2021.

It’s critical for IT organizations to have more flexible IT infrastructure platforms. One of the primary reasons so many application workloads are moving to public clouds is that the internal IT organization is considered too slow to meet the needs of developers of modern applications. Instead of waiting weeks or months for an internal IT organization to provision all the compute, storage and networking resources developers may want to invoke on demand, those developers are turning to cloud service providers that give them access to those resources in a matter of minutes.

The good news for many internal IT infrastructure teams is that there are still workloads that are either deemed to be too latency sensitive to be deployed on cloud services that need to be accessed over a wide area network (WAN) or are not eligible to be deployed outside of a data center controlled by the internal IT organization because of compliance issues.