Kublr, a provider of a management platform for Kubernetes clusters, has teamed with cloudtamer.io to help IT organizations rein in costs stemming from deploying Kubernetes clusters on public clouds.
As IT organizations move to operationalize Kubernetes clusters, many of them are discovering that dynamic attributes that made the platform popular with developers make it difficult to predict costs. The governance tools developed by cloudtamer.io make it possible to define policies based on the budget dollars allocated to any Kubernetes deployment.
Company CEO Brian Price says in the wake of the economic downturn brought on by the COVID-19 pandemic, there’s more sensitivity to cloud costs than ever. The challenge is finding a way to proactively rein in those before they receive their monthly cloud service provider bill.
Kublr CTO Oleg Chunikhin adds governance tools such as cloudtamer.io are a natural extension of the company’s namesake platform for managing Kubernetes clusters. The cloudtamer.io governance platform surfaces cost insights that inform the Kubernetes management plane provided by Kublr without having to exit the platform, he says.
Aimed primarily at IT administrators, Kublr presents a graphical user interface (GUI) through which IT teams can manage fleets of Kubernetes clusters. The goal is to reduce the total cost of managing multiple clusters at a time when multiple application owners inside an organization are starting to deploy Kubernetes clusters. In theory, each of those application owners could share a Kubernetes cluster. In practice, each application owner typically prefers to spin up their own Kubernetes cluster. Before too long organizations can find themselves managing multiple Kubernetes clusters that at different times are driving spikes in consumption of cloud resources.
In fact, it’s the unpredictability of those costs that is drawing more scrutiny from finance teams. Organizations are clearly vested in building and deploying cloud-native applications built using containers to drive, for example, digital business transformation initiatives. However, in an era marked by tight budgets, tolerance for surprise IT expenses is near-zero—the days when DevOps teams could count on a blank cloud check are now over. Finance teams are also demanding a lot more visibility into which cloud service providers cost more to employ than others. The preference of a DevOps team may not count for much when there is an existing agreement in place that requires an organization to consume a specific amount of resources by the end of the year to qualify for an annual discount.
It’s not clear to what degree organizations will rely on traditional IT administrators to manage Kubernetes clusters versus DevOps teams that tend to prefer application programming interfaces (APIs) and command-line interfaces (CLIs). Regardless of who is in charge, finance teams are holding IT organizations more accountable for costs.
Obviously, there may be a legitimate reason why a spike in activity drove up IT costs. However, uncomfortable questions will be asked if costs are rising at a time when there is excess Kubernetes capacity available on a spot cloud market. The IT team that isn’t able to immediately explain why those increased costs were unavoidable is going to be viewed as being a lot less competent than IT teams that can.