Join top executives in San Francisco on July 11-12, to hear how leaders are integrating and optimizing AI investments for success. Learn More
Driven by a need for a faster and smoother software development process, the rising adoption rate of cloud-native technologies creates a massive knowledge gap between technical and non-technical teams.
Finance departments struggle to understand the cost dynamics of cloud computing. And modern cloud-native approaches like Kubernetes step up the challenge around cost allocation and management.
The State of FinOps survey showed that getting engineers to act on cost optimization recommendations is a top challenge for nearly 40% of respondents, no matter their maturity level.
Why are Kubernetes costs so hard to understand?
Before containerization, allocating resources and costs was more straightforward. All it took was tagging resources to a particular project or team, and the finance team would get all the data for identifying cost structure and controlling the budget.
Join us in San Francisco on July 11-12, where top executives will share how they have integrated and optimized AI investments for success and avoided common pitfalls.
As Kubernetes and other containerization solutions became more widespread, the traditional process of allocating and reporting on costs failed to do its job due to challenges around shared resources and Kubernetes-specific resource utilization.
Still, engineering teams need finance’s buy-in to benefit from cloud solutions that push developmental efficiency to new heights and increase business agility.
FinOps is an approach that addresses that very challenge by offering a cluster of best practices applicable to every part of the organization. How can organizations take advantage of FinOps and spread awareness of cloud costs among both technical and business teams?
Implement FinOps to put engineering and finance teams on the same page
The following steps draw from FinOps best practices and allow technical and busiest teams to find common ground in cloud cost management:
Establish a common platform for cost visibility
Ideally, the cloud cost monitoring solution caters to the needs of both teams. It generates reports that are understandable to finance and exposes metrics that are easy to grasp for engineers. Ideally, these metrics are scrapable through tools like Prometheus and can be added to dashboards in monitoring solutions engineers already use, such as Grafana.
That way, engineers aren’t forced to switch contexts and work with another tool on top of the dozens they use just to check how much their Kubernetes cluster costs.
Use historical cost data for fixing issues and budgeting
A recent survey revealed that cloud cost issues could cause serious disruptions to engineers’ work: 41% of respondents said cost problems cause interruptions that last at least a few hours per week. For 11%, cloud costs led to high interruption equivalent to a sprint or greater.
Many of those teams have no access to historical cluster cost data. So, if an incident happens in this context, it’s pretty realistic for a team to spend one sprint or more investigating where that sudden cost spike came from. Implementing a cost monitoring solution with access to historical cost data shrinks the investigation time to minutes, giving all teams access to granular cost data.
Moreover, by giving both teams access to this data, planning becomes a common effort for engineering and finance. A good cost monitoring solution offers a glimpse into historical spend and shows the daily level of cloud expenses to help engineers keep track of the cloud budgets they have set together with finance.
Provide access to real-time cost data
This point is tricky as none of the major cloud providers offer access to cost reports generated in real time. Third-party solutions that increase cost visibility fill this gap, allowing engineering teams to instantly identify cost spikes and keep their cloud expenses in check.
This is especially important since engineers don’t have the time to constantly keep an eye on the infrastructure. At the same time, organizations need to protect themselves against the risk of, for example, leaving a job running for longer than it should and ending up with a surprise cloud bill of over $500k, as Adobe did. One alert acting on real-time usage and cost data can prevent this.
Prepare for FinOps 2.0
While Finops is a relatively new term, the practice of monitoring and reporting on cloud expenses likely emerged together with the spread of public cloud services. Companies that jumped on the cloud bandwagon soon found that while a cloud migration might save them data center costs, it also comes with a wide range of new financial challenges.
To control cloud costs, companies used various cost monitoring, reporting, and allocation solutions that relied on manual tasks such as meticulous resource tagging.
There’s no reason why FinOps should continue this way. Automation tools are solving so many problems in the industry already, so why not use them in this space? After all, the ultimate goal of FinOps is controlling and reducing cloud costs. Cost optimization solutions that rely on automation can bring teams to that stage in a matter of minutes.
So, here’s another proven best practice that puts the finance and engineering teams on the same page:
Leverage automated cloud cost optimization
Demand and utilization change rapidly in cloud-based applications, and managing costs manually quickly becomes time- and labor-consuming.
Solutions that automate tasks such as provisioning new virtual machines, finding the best match for the application requirements, or replacing interrupted spot instances with new ones help teams to achieve financial goals without the added effort from the engineering side of the organization.
Building a FinOps culture starts with collaboration between engineering and the finance team
Running Kubernetes in the dark is risky. In the worst-case scenario, it results in a snowball effect where the organization has no idea which applications, services, or teams consume cloud resources and generate costs — and how these translate into the budget the finance team set for the month.
To create a strong FinOps culture where both engineering and business teams understand and take ownership of cloud costs, organizations need to help these teams find common ground. That’s because cost data that make sense to finance may not resonate with engineers and vice versa.
By equipping teams with a platform that delivers cost insights in the right format and location — be it a financial report or dashboard in a popular monitoring tool — organizations can take the first step to keep their costs under control.
Laurent Gil is cofounder and chief product officer at CAST.AI. He formerly led Oracle’s internet intelligence group.
Welcome to the VentureBeat community!
DataDecisionMakers is where experts, including the technical people doing data work, can share data-related insights and innovation.
If you want to read about cutting-edge ideas and up-to-date information, best practices, and the future of data and data tech, join us at DataDecisionMakers.
You might even consider contributing an article of your own!