While many enterprises still see the cloud as a useful way of providing additional capacity over the next few years, they’ll come to see it as a source of innovation and by 2028 most will see it as a necessity, according to Gartner analysts.
Part of that will be the result of the conversion or rewriting of legacy applications — perhaps running on mainframes or legacy mid-range systems — to make them ready for the cloud age. At its recent IT Infrastructure, Operations & Cloud Strategies conference in London, analyst firm Gartner said that by 2028, these modernization efforts will result in 70% of tech workloads running in a cloud environment, up from just 25% today.
Jump to:
- Gartner analyst: Why businesses will need cloud to survive
- Tips for monitoring and controlling cloud computing spending
- How to improve cloud computing resilience
Gartner analyst: Why businesses will need cloud to survive
Dennis Smith, distinguished vice president analyst at Gartner, said that by 2028 cloud computing won’t just be needed for businesses to be competitive, but for their survival (Figure A). But moving the more business-critical applications to the cloud — the ones that haven’t been moved so far — is likely to be complicated, he told TechRepublic during a Zoom interview.
Figure A
“A lot of the easy workload has moved and now comes a lot of the tough stuff, the critical stuff,” he said. “So I think the big question is ‘Where does it go and what will it look like?’ Is it a matter of it migrating to a public cloud or does it mean that cloud-related technologies will come to me, within my on-premises environment?”
New cloud computing choices for business
Smith said that shifting critical apps to a cloud model might not mean shifting them to a public cloud data center. Rather, it could mean imposing cloud attributes on those workloads, even as they remain in corporate data centers. That might mean introducing cloud-inspired concepts such as automation, self-service and consumption-based usage, he said.
Gartner said there is also going to be more interest in industry cloud platforms; these combine the underlying cloud services with a set of modular industry-specific offerings. The analyst firm said that by 2028, more than 50% of enterprises will use industry cloud platforms to accelerate their business initiatives.
SEE: Cloud Platform Services: Quick Glossary (TechRepublic Premium)
And while much of the attention around cloud is focused on a few big platforms, there are likely to be many more options to choose from.
“You are going to see an expanding of the cloud ecosystem. When we think of cloud, we think of the handful of hyperscalers but there’s a tonne of other vendors out there,” he said, providing everything from management tools to database and application tools.
Smith said there will continue to be a focus on operational aspects like cost governance: “Companies are now recognizing that cloud spend can be significant and can easily get out of control if they aren’t managing it,” he said.
How cloud computing priorities are changing
But some issues are less dominant now. “Cloud computing is becoming the de facto security model and most people would accept that most cloud deployments are more secure than what you would have in a more traditional data center,” he said.
However, an area of growing interest will be digital sovereignty, which aims to ensure that cloud data is being held under the correct data protection regime. By 2028, over 50% of multinational enterprises will have digital sovereignty strategies, up from less than 10% now, said Gartner (Figure B).
Figure B
For companies planning to join this new wave of cloud computing, a vital step is to audit the systems that they still have in-house.
“If there’s one fundamental area for enterprises, it is that you need to go house-by-house or closet-by-closet to really look at your application portfolio and decide what is best to do with it, what can be replaced what needs to be replatformed and what needs to be refactored,” Smith said.
Getting staffing right is also key, he said. With the cloud becoming so important, it’s no longer sufficient to have a parallel organizational structure with a cloud team that’s separate from the legacy team with the hope that over time they will mingle or work out, he said. “If cloud is becoming an integral part of your technology or business strategy, then you need a plan for that.”
SEE: Hiring Kit: Cloud Engineer (TechRepublic Premium)
Tips for monitoring and controlling cloud computing spending
However, even if cloud migration is where the momentum is, execs should still make sure they’re able to monitor and control cloud spending to avoid overspending.
In his presentation on day two of the event, Gartner VP analyst Paul Delory set out how to evaluate public cloud costs and implement best practices, warning that simply shifting something to the cloud alone isn’t going to cut costs. “You must invest time, effort and skills to ensure your cloud usage is efficient,” he said in a press release.
Among Delory’s recommendations, which are listed in the press release:
- Don’t use cost as your biggest factor in cloud adoption; set it alongside core drivers of agility, flexibility and innovation.
- Don’t assume; calculate the total cost of ownership and return on investment.
- Evaluate any shift over the long term — such as five years — because benefits and return on investment may not be realized in the short term.
- Execs should focus investment on the most impactful activities, such as those with the highest potential savings as 80% of the savings are often achieved with only 20% of the time investment.
How to improve cloud computing resilience
While cloud outages tend to be rare, they do happen. At the same event, Gartner provided guidance to improve cloud resilience. Its recommendations include:
- Align resilience requirements to business needs so that teams don’t fall short of resilience expectations — or overspend.
- Take a risk-based approach that puts emphasis on common failures that organizations have greater control over.
- Build dependency graphs that map all middleware components, databases and cloud services so they can be configured for resilience.
- Implement automated disaster recovery to provide the foundation for aggressive recovery time objectives.
