Rafay Systems Adds Kubernetes Cost Optimization Capabilities to Management Platform
Rafay Systems today added a cost optimization suite of tools that continuously analyzes Kubernetes costs in a way that enables IT teams to invoke its management platform to automatically rein them in.
Additionally, the company published the results of two surveys of more than 2,000 IT professionals from organizations with more than 1,000 employees, that found managing cost visibility and controlling costs is currently the top challenge IT teams managing Kubernetes clusters face, followed by complexity (38%) and maintaining standardization (38%).
A full 60% said reducing and optimizing costs associated with Kubernetes infrastructure remains a top management initiative for the next year. Nearly a third (31%) said the total cost of Kubernetes ownership, including software/support licenses and salaries, is higher than budgeted for or anticipated.
Rafay Systems CEO Haseeb Budhani said when it comes to controlling costs organizations typically lack the expertise required to dynamically scale Kubernetes clusters up and down as needed. The cost optimization suite provided by Rafay makes it easier for IT professionals of any skill level to automatically apply policies to, for example, automatically rightsize clusters to better control costs using a set of tools that abstract away the underlying infrastructure complexity, he added.
IT teams can also track historical consumption by workloads and teams to enable chargeback or show back of infrastructure consumption.
These are critical capabilities when organizations are looking to build and deploy more cloud-native applications using platform engineering teams that are employing best DevOps practices to manage multi-tenant Kubernetes environments at scale, noted Budhani. Unfortunately, too many IT teams are trying to use legacy approaches to managing virtual machines to Kubernetes clusters, which results in increased cost as infrastructure resources are over-provisioned, he added.
The survey finds the top priorities for improving developer productivity are automating cluster provisioning (47%), standardizing and automating infrastructure (44%), providing self-service experiences for developers (44%), automating Kubernetes cluster lifecycle management (44%) and reducing cognitive load across developer teams (37%).
Achieving those goals should become simpler as management platforms such as Rafay continue to add generative artificial intelligence (AI) capabilities that will help democratize the management of Kubernetes clusters, noted Budhani.
At the same time, Kubernetes itself is rapidly becoming the default platform for deploying artificial intelligence (AI) applications. The challenge is that 94% of the 1,035 respondents with AI application experience admit they are struggling with generative AI application experimentation and deployment.
Less than a quarter of organizations have a sufficient level of implementation for both machine learning (MLOps) (17%) and large language model operations (LLMOps) (16%) workflows defined. Specific challenges include security for MLOps and LLMOps workflows (50%), model deployment automation (49%) and data pipeline management (45%).
It’s not clear to what degree AI will ultimately drive increased consumption of Kubernetes clusters, but as the pace at which these applications are built and deployed continues to increase, so too will the number of IT teams that are trying to effectively manage these environments. The challenge, as always, will be making it possible to automate the management of Kubernetes clusters at scale in a way organizations can afford.