Anyone who has been in IT for more than a few years knows the headache of legacy infrastructure. As equipment and software age, they become less capable of supporting new applications, more prone to failure, and vulnerable to security breaches as well as loss of vendor support.
The problems only get worse with time. Software doesn’t live in a vacuum; it must communicate with other programs, and the requisite tools can lock businesses into an aging platform for the long term.
“We often find that customers that don’t invest in their legacy applications end up running older versions of middleware, which can, in turn, require older versions of operating system software and thus older hardware,” says Stanley Wood, IBM distinguished engineer and chief technology officer for Kyndryl’s Managed Private Cloud Infrastructure-as-a-Service (Managed Private Cloud IaaS) offerings.1
“It can be become an expensive mess to keep it running at all while security risks grow,” especially when older commercial applications age out of support contracts, he adds.
Old applications and infrastructure can also drag down the scalability of newer platforms, by restricting capacity. “The underpinnings of the application are not designed to support the size of the workload,” Wood says. Legacy systems thus effectively become the weak link in modern infrastructure.
Cloud platforms offer relief from worries about infrastructure currency or capacity. The cloud provider “ensures there’s extra capacity on hand while customers pay only for what they use,” Wood says.
But migrating to public cloud platforms can be expensive and risky, particularly for businesses in regulated industries such as financial services, healthcare, or the public sector. Removing applications from production in the data center can cause dependent software to fail. And latency issues between on-premises infrastructure and public cloud can complicate time-dependent processing that was built with the assumption that all elements would exist in proximity.
“We’ve helped a lot of clients move to the cloud and then move back in a big hurry because of these considerations,” Wood says. An early failure can leave organizations gun-shy about future migration projects.
There is a better way: Kyndryl’s Managed Private Cloud IaaS takes the burden of maintaining, upgrading, and patching legacy infrastructure off the IT organization and shifts it to a managed service provider. Customers get benefits of a modern platform that are equivalent or even superior to those of a public cloud, with none of the management overhead. New infrastructure arrives fully tested and integrated. Failover to public infrastructure can be made automatic and transparent.
Managed private cloud infrastructure also addresses the very real problem of skills development. Maintaining physical data center infrastructure is different from managing a highly automated software-defined cloud environment. Although most data center staffers are more than capable of making the transition over time, it can take a year or more to get a self-hosted private cloud up and running. A managed service provider can deliver rapid results and give an IT organization time to develop skills on its own schedule.
“It’s hard to pivot from the old way of looking at infrastructure to the new,” Wood says. “By outsourcing to a provider, you don’t need to entirely reshape the environment, just shift it.”
A managed private cloud is the future-proof infrastructure that has long eluded IT organizations. “The customer doesn’t have to worry about the currency of the hardware or systems software,” Wood says. That’s one worry that CIOs who are busy ushering their organizations into a digital future will be happy to give up.
For more information on IBM’s Managed Private Cloud IaaS, visit ibm.biz/PC_IaaS.
1Kyndryl is currently a wholly-owned subsidiary of International Business Machines Corporation with the intent that Kyndryl will be spun-out.