One of the enduring debates in the digital era is how best to ensure cloud bills do not run higher than the finance team is comfortable with.
The technology increasingly forms the “backbone” of modern enterprise but that then creates inevitably rising usage that must be optimised.
Smart companies using cloud in the right way, however, can reap the benefits of modernisation while ensuring their usage is managed strategically.
As Grant Gross observes in CIO.com: “Cloud market observers point to large AI workloads driving up spending, but also to developers who aren’t clued-in to the cost of the cloud services they consume.”
Capacity or cost: the cloud challenge
The cloud’s flexible, on-demand nature allows IT teams and developers to “spin up” or deploy resources with little more than an online account and a credit card. But this flexibility means that cloud spending can bypass conventional IT cost controls.
At the same time, short-term cloud capacity, from development servers to data storage, can all too easily become a long-term commitment. Without the right management, businesses can find they are paying for unused resources, or for the wrong tiers of compute and storage. AI workloads, with their extensive demand for cloud resources, add to the problem.
Research by the management consulting firm McKinsey found that most organisations could save 10 to 20% of their cloud spending. Automating cloud cost management could boost savings further still. But CIOs need to drive efficiencies, without limiting innovation.
In some ways, this is becoming harder. Cloud pricing can be complex. Costs fluctuate, and organisations face different charging models across geographies. Usage-based pricing means costs can creep up, and there is a temptation to over-provision services “just in case.”
Some providers also levy charges for services, such as data egress, which can catch users out. Older versions of cloud services can also be more expensive than the latest technology.
And the specialist nature of AI services, such as graphics processing unit hours, adds to costs, as does the rapidly rising demand caused by generative AI and large language models.
Bring on finops
Technology leaders can, however, balance innovation with the need to control costs. Cloud financial management, also known as “finops,” is increasingly important to ensure that the business has access to the technology it needs, at a cost it can afford.
Cloud vendors are aware of the challenge and have invested in cost management. AWS, for example, has a suite of tools to monitor usage, for budgeting, forecasting and to optimise resource utilisation.
AWS Cost Explorer, for example, improves awareness around cloud spending. AWS Instance Scheduler allows firms to turn capacity on and off, based on predicted demand, and AWS Cost Anomaly Detection alerts users to unexpected changes in their cloud spending.
In addition, enterprises can save more by changing the way they buy cloud services and by automating finops. McKinsey, for example, predicts that enterprises globally could save $120 billion, out of a spend of $440 billion, on infrastructure and platform-as-a-service costs through automating finops or “finops as code.”
And AWS itself says that its customers can save up to 72%, compared to on-demand prices, by moving to its Savings Plans. And those savings can go directly back into supporting innovation across the business.