While cloud-native technologies are opening a path to greater productivity, organizations must overcome technical management and operational challenges to realize that potential.
Meanwhile, cloud leaders should be focused on refining their strategies and picking up the pace of adoption.
These were among the findings of Forrester Research's recent "Top 10 Trends in Cloud" report.
How Cloud Can Redefine the Data Center
The study also pointed to the continued dominance of hyperscalers — aka the cloud mega-vendors AWS, Microsoft Azure, and Google Cloud — and predicted cloud technology's potential to redefine the data center.
"The virtualization of the enterprise data center in the last two decades has yielded tremendous gains in productivity," Forrester principal analyst Lee Sustar said. "However, the technology underlying current virtual machines is aging, brittle and cannot support application development and deployment at scale that is needed in the hyperscaler era."
Nick Durkin, field CTO for Harness, explained that the cloud is redefining the enterprise data center by minimizing the data center's value, or in many cases making data centers entirely obsolete.
"Shareholders do not pay more for the stock of a company because they have built and maintained highly available, fully fault-tolerant data centers," he said. "Their ability to handle power outages, network interruptions, fire, or natural disasters at their data center no longer increases their value because the cloud providers have commoditized this resiliency."
As part of the effort to accelerate cloud-native adoption strategies, Durkin pointed out that the business is the key stakeholder, along with the CFO organization.
"The arguments for modernization are compelling in the abstract but have to be related to particular environments and approached in a strategic fashion," Sustar said.
Who's on the Front Lines of Cloud Operations?
The group on the front lines of cloud-native adoption is, of course, the people who will be using these tools daily: developers and engineers, Durkin said.
"When given the information, engineers do what is best for the business," he added. "Denying them information like cloud costs throughout the entire SDLC truly hurts the business."
If an engineer was shown irrefutable evidence that they have overprovisioned infrastructure even at peak load, they will gladly right-size that infrastructure, he said.
"Withholding access to data that can increase efficiencies in both cost and performance would never be done in any other part of the business," Durkin explained. "Cloud cost and efficiencies should start in the very beginning of the SDLC, not after it's already running."
Cloud hyperscalers will likely continue to demand an increasingly large share of the IT budget, Sustar said.
"The hyperscalers' sheer size and their ability to mediate the flow of data and transactions at global scale give them a commanding position in shaping the future of IT," he said.
While cloud customers have appreciated the rapid rate of innovation and proliferation of services, they now want the hyperscalers to provide more automation and finished or semi-finished platforms that are easier to adopt, Sustar said.
In addition, the distinction between cloud and data center is breaking down as Kubernetes-based multicloud and hybrid solutions provide a common abstraction layer, as do various "cloud at customer" solutions from cloud providers.
Even hyperscalers struggle to stay up on the latest advancements in technology, Durkin pointed out.
"Like all businesses, they have to shift and evolve as the technology does," he said. "Often hyperfocus can be an advantage; other times it tends to act like blinders. Making sure you are always staying two steps ahead of your customers will guarantee you are always prepared for the challenges to come."
Managing Cost Is Critical to Cloud Operations
Cost management will be a critical aspect of any strategy going forward, according to Sustar.
"Organizations spent a lot of money on cloud services in a short period of time to alleviate pressures during the worst of the pandemic," he said. "A period of rationalization was bound to follow, and economic uncertainty will make that process more intense."
From Durkin's perspective, rising cloud adoption brings an inevitable rise in cloud spend.
"This creates a crucial problem for teams — estimating and budgeting cloud spend for various operations, and reducing cloud COGS for SaaS applications that drive a company's revenue," he said.
Durkin said any cost a company incurs in this economic climate should not just be managed but governed.
"Make it easy for engineers to do the right thing and make it hard for them to do the wrong thing," he said. "Shifting the information left should be the No. 1 priority."
In addition, cost information should be exposed, tracked, and optimized from the very beginning of the SDLC.
Engineers should have the same access to costing information, budgets, and savings potentials that finance teams have, he said.
"Engineers will make decisions that benefit the company by default, but can only do it based on the data they have, and cost data has been almost nonexistent," Durkin explained.
About the authorNathan Eddy is a freelance writer for ITPro Today. He has written for Popular Mechanics, Sales & Marketing Management Magazine, FierceMarkets, and CRN, among others. In 2012 he made his first documentary film, The Absent Column. He currently lives in Berlin.