Cloud is powerful. We know that cloud computing services are capable of delivering flexibly controllable computing processing, data analytics and storage power in potentially huge quantities. But because the cloud is powerful, it also takes up a lot of power. That power comes in the form of the electricity required to run the server estates inside today’s Cloud Service Provider (CSP) datacenters, it is a power that comes in the form of data throughput in and out of the cloud as organizations and users plug into the backbone therein – and it is power in terms of cost for all of the above, in cold hard cash.
Controlling the cost of cloud – or at least accurately auditing and assessing the cost of cloud on the road to trying to quantify, qualify and control it – is typically now associated with the practice of FinOps. A discipline devoted to the financial analytics of cloud services with a view to financial management, CloudBolt Software styles itself as the cloud ROI company (a term it has trademarked™ with a view to being perceived as more than ‘just’ FinOps) as interest in this domain now grows, especially as we now embrace Environmental, Social & Governance (ESG) concerns.
Kubernetes: almost de facto status
Already recognized for its own cloud financial management platform and toolset of technologies, CloudBolt has now cemented a partnership with StormForge, a provider of Kubernetes resource optimization technologies. As previously explained here, Kubernetes (sometimes referred to as “K8s”) is an open source orchestration system used to streamline the deployment, scaling and management of containerized software applications and services. Because Kubernetes has become so widely embraced by the cloud engineering cognoscenti, its almost de facto status warrants it a special focus across both new greenfield and existing brownfield technology projects looking to operate with more composable efficiency. Seeking out new Kubernetes capabilities to add to its cadre of cost control capabilities, CloudBolt’s union with StormForge is hoped to bring Kubernetes cost visibility and optimization into the FinOps ecosystem.
According to a survey by the Cloud Native Computing Foundation (CNCF), Kubernetes is used in 96% of global businesses. Public cloud adoption continues to accelerate, and the enterprise ‘lift and shift’ pattern for cloud migrations is being replaced with a focus on modernization. Yet, 69% of organizations either do not monitor Kubernetes spending at all or rely on monthly estimates.
While the FinOps industry has had success providing cost visibility and optimization recommendations at the Infrastructure-as-a-Service (IaaS) layer, it has not yet delivered the same value in Kubernetes ecosystems. The pervasive target for all new workloads is therefore a cost-efficiency black box and as a result, monthly Kubernetes costs are sometimes argued to be capable of spiralling into tens or hundreds of thousands of dollars for medium to large enterprises.
Without any form of monitoring, organizations remain unaware of actual spend and over-allocated cloud resources. Because we know that calculating Kubernetes costs using manual methods is time-consuming and often inaccurate, organizations often find that can provide little insight into the root causes behind cost increases. As a result, platform and FinOps teams are siloed and starved of insights that would inform meaningful action.
CloudBolt says that the answer, it hopes, is the fact that StormForge’s intelligent machine learning capabilities will now be integrated with CloudBolt’s Augmented FinOps offerings to deliver full Kubernetes cost visibility with detailed cost allocation against real-time performance metrics and fully automated Kubernetes optimization.
Floundering foundational FinOps functions
“As Kubernetes adoption continues to accelerate, costs have exploded and foundational FinOps capabilities like chargeback and container level visibility prove to be difficult or absent altogether, and rarely result in optimization,” said Kyle Campos, CloudBolt’s chief technology and product officer. “This strategic partnership will empower Kubernetes users with Machine Learning (ML)-based FinOps capabilities to effectively manage Kubernetes expenditure by not only delivering cost visibility, allocation and rightsizing recommendations, but also natively orchestrating the execution of those recommendations against Kubernetes clusters.
“Kubernetes users have not had a path to maximize their cloud ROI with the market’s lack of a cohesive solution to inform, optimize, and automate Kubernetes cost reporting. We’re excited to close that loop by seamlessly integrating our ML-driven, automated Kubernetes optimization with CloudBolt’s Augmented FinOps platform,” said Yasmin Rajabi, StormForge’s VP of product.
The joint technology offering here provides complete shared cost distribution. This means CloudOps teams can now align with FinOps teams for showback/chargeback that accurately allocates costs based on real container-level performance data to create non-contested attribution. With a process known as continuous rightsizing, cloud practitioners now have the ability to reduce insight-to-action lead time through continuous Kubernetes optimization. For improved cloud ROI, Kubernetes clusters are included into cost visibility and optimization workflows that have previously been isolated from the rest of cloud spend.
Despite the rise of automation solutions designed to make cloud more simple to run, use and pay for, cloud computing is generally considered to be increasingly charged with complexity and convoluted configuration headaches. Putting a more orchestrated financial management consideration into the freshly boiling stewpot that is Kubernetes will be welcome news for some. When we start talking about Kubernetes cost as ‘kost’, then you’ll know it’s time to cash in.