How tech suppliers are easing barriers to hybrid cloud adoption

Enterprise cloud adoption is growing steadily across the Asia-Pacific region, with most companies using some form of software as a service (SaaS) or running selected workloads on a public cloud infrastructure.

Yet, even as some tech suppliers believe that, in the fullness of time, all enterprise workloads will move to the public cloud, the reality – at least for now – is the preference for a hybrid cloud set up for various reasons, such as performance and compliance requirements.

That requires enterprises to be able to run and manage their workloads consistently anywhere, a message that IBM has been espousing since it acquired Red Hat, whose OpenShift container platform lets organisations run containerised applications on multiple public clouds and on-premise.

Jim Freeman, IBM’s new chief technology officer for global technology services in Asia-Pacific, said enterprises that operate in a hybrid brownfield environment are also seeing an average of 2.5 times more business value than cloud-native firms, citing recent studies.

This trend has not gone unnoticed by cloud suppliers, many of which have risen to the challenge of helping enterprises ease their transition to cloud computing.

“The first thing is they’re providing increased transparency, particularly to be compliant with regulatory and other requirements,” said Freeman. “It’s one thing to just say trust me, but it’s another to open up your log files and your procedures to demonstrate compliance. That is increasingly recognised as table stakes for enterprise cloud adoption.”

Also, Freeman said data security is critical, so enterprises are assured that their data is safe from prying eyes, noting that IBM Cloud has made it “technologically impossible” to look at its customers’ data. “That helps not only the compliance people, but also the CIOs and their clients,” he said.

With the multitude of cloud services available, the industry is also stepping up efforts to create an integration framework to ease the demands of managing all the integration points.

“I’m pleased to say that the industry is kind of saying: let’s build an integration framework that enterprises want to consume and ask all our ecosystem partners to conform to that framework,” said Freeman. “That demonstrates the maturation of enterprise cloud.”

He said some standards have also started to emerge around managing enterprise services in a heterogeneous environment, paving the way for wider enterprise cloud adoption. These could be single sign-on and backup and recovery services, so security and data management can be coordinated across on-premise environments and public clouds, he added.

Even with these standards, enterprises could still be locked into specific cloud platforms if their applications rely on proprietary cloud services, particularly those around artificial intelligence (AI) and machine learning – going against the notion of building applications and running them everywhere.

To that, Freeman said standard services interfaces have started to emerge around “base level” commodity services such as databases, but more sophisticated offerings such as AI will continue to be prized by cloud suppliers as competitive differentiators, making them unlikely to be commoditised for a while.

Some enterprises have started to consider moving workloads from public cloud to on-premise datacentres in what is now known as cloud repatriation, but that is often a result of a less robust workload placement strategy and the lack of cloud management skills, said Freeman.

He said enterprises should have a cloud workload strategy that identifies the functional requirements for an application. Equally important, he added, are non-functional requirements such as data gravity, or the ability of data to attract applications, services and other data.

“Cloud repatriation is often the result of two things: a lack of understanding that you should be turning things off when you’re not using them, and the failure to realise the data consumption rate of an application,” said Freeman.

But how should enterprises decide on their workload placement strategy? Freeman advised them to start simply with what data gravity, security and CPU consumption would look like. “With those three things, I can give you a score of your target platforms pretty quickly,” he said.