The FinOps Foundation today made available an update to a specification for controlling IT costs that adds an ability to split shared resource costs, track contract commitments accurately, and verify data freshness.
Version 1.3 of the FinOps Open Cost and Usage Specification (FOCUS) specifically makes it possible to allocate shared costs for containers and databases, adds an external dataset for contract or negotiated agreements, and provides access to timestamped metadata that makes it possible to determine completeness and recency of signals collected.
Additionally, definitions will be more delineated between service and host providers to make it easier for IT to more easily see who to engage for support and billing inquiries.
Matt Cowsert, principal product manager for FOCUS, said the goal is to make it easier for IT teams to move beyond spreadsheets to track and control IT costs. The ability to allocate shared costs is especially useful in organizations that need transparency into how multiple departments are consuming the same set of core cloud services, he added.
That issue is only going to become more important as data-intensive artificial intelligence (AI) workloads are added to the portfolio of applications that IT teams need to manage, noted Cowsert.
FOCUS is already supported by Alibaba, Amazon Web Services (AWS), Databricks, Google Cloud, Grafana, Huawei, Microsoft, Oracle, OVH Cloud, and Tencent, and more organizations in recent years have adopted a set of best FinOps practices to better control IT costs, especially cloud computing costs that tend to fluctuate from one month to the next. While many organizations would ideally like to reduce those costs, the first priority is usually trying to gain enough visibility to be better able to predict the cost of cloud computing services before the monthly bill arrives.
That can be especially challenging because application developers for years now have been using corporate credit cards to provision cloud infrastructure resources with little to no oversight. Unfortunately, developers are also prone to leaving the proverbial lights on by, for example, leaving application development environments running over a weekend or simply forgetting that a virtual machine has been left to run for weeks—sometimes even months.
Developers also often ignore alerts signaling possible spikes in consumption due to, for example, a data-intensive workload unexpectedly consuming more infrastructure resources.
It’s not clear precisely how organizations are implementing best FinOps practices. Some have created dedicated teams made up of IT and finance professionals that have deployed tools that provide insights into both cloud computing and software-as-a-service (SaaS) application costs, while others are simply trying to make it easier for application developers to track cost along with other performance metrics they already monitor.
Regardless of approach, the need to better control costs is only going to become more pressing as the number of workloads deployed in cloud computing environments continues to increase. Of course, no one can manage what they can’t see so the first step is often simply determining how many workloads are running exactly where.
