Select Page
Affiliate Disclosure: This page may contain affiliate links. When you click and make a purchase, we may receive a commission at no additional cost to you. Thanks for supporting our content.

Solving the Kubernetes Autoscaling Challenge The Easy Way

Sep 29, 2022 | Episode

Solving the Kubernetes autoscaling challenge. Find out about Bi-dimensional POD autoscaling. Scott Moore interviews Patrick Bergstrom, Chief Technology Officer of StormForge.

Kubernetes Autoscaling is a significant challenge (and opportunity) for Kubernetes users. Out of the box, Kubernetes comes with two primary ways of dynamically scaling – the Horizontal Pod Autoscaler (HPA) and the Vertical Pod Autoscaler (VPA) – but it’s not possible to use both together without extensive effort and customization. Additionally, the HPA requires users to configure a target utilization used to determine when to scale, but setting that target utilization in an optimal way is a challenge.

Scott Moore interviews Patrick Bergstrom, Chief Technology Officer of StormForge. They discuss the new release of their StormForge Optimize Live product, and how advanced machine learning solves the problem of using horizontal and vertical auto scaling on the same workloads for Kubernetes deployed applications. It works at both the node and container level, replacing guesswork with science to eliminate the choice between cost and performance.

Video Insights On Kubernetes Autoscaling

🚀 StormForge Optimize Live uses machine learning to automatically right-size Kubernetes components in real-time, leveraging observability data to make smart decisions on pod sizes for workloads.

🔄 The platform enables bi-directional scaling by combining Vertical Pod Autoscaler (VPA) and Horizontal Pod Autoscaler (HPA), maximizing compute efficiency and reducing waste in Kubernetes environments.

Cloud Efficiency

💰 StormForge’s solution addresses cloud waste in large Kubernetes clusters with cookie-cutter resource deployments, providing automated, ongoing production recommendations to optimize efficiency for platform operators.

🧠 The platform’s machine learning-based right-sizing helps solve the D2 complexity gap in Kubernetes, enabling organizations to fully leverage its capabilities without requiring extensive manual intervention.

Leadership and Performance Engineering

👥 A servant leadership culture focused on removing roadblocks and supporting staff success is crucial for retaining top performers and preventing stagnation in engineering teams.

🔬Performance engineering is often overlooked by senior leaders, but StormForge’s automated approach can help organizations improve cloud spend efficiency while increasing application performance.

StormForge Unveils Bi-Directional for Kubernetes Autoscaling

StormForge, a company specializing in AI-powered solutions for Kubernetes, has announced the launch of their groundbreaking bi-directional scaling feature.

The Challenge of Kubernetes Autoscaling

Scaling applications deployed on Kubernetes efficiently is a significant challenge for many organizations. Over-provisioning resources leads to unnecessary cloud costs, while under-provisioning can result in performance issues. Traditionally, Kubernetes users have relied on two separate tools to handle scaling: the Vertical Pod Autoscaler (VPA) and the Horizontal Pod Autoscaler (HPA).

  • The VPA adjusts the size of individual pods (units of deployment in Kubernetes) based on CPU and memory consumption.
  • The HPA increases or decreases the number of pod replicas based on specific thresholds.

However, using the VPA and HPA together often leads to “thrashing,” where they compete with each other to solve scaling issues. This forces organizations to choose between the two, limiting their scaling flexibility.

StormForge’s Solution: Bi-Directional Scaling

StormForge’s new bi-directional scaling feature eliminates this dilemma by allowing both the VPA and HPA to work in harmony. This is achieved through a sophisticated algorithm that optimizes pod sizing in conjunction with HPA replica adjustments. This approach ensures efficient resource utilization and avoids thrashing, maximizing both performance and cost savings.

Bergstrom highlighted that StormForge’s solution is the first of its kind in the market. He emphasized the potential cost savings, estimating that organizations could reduce their cloud bills by up to 30% while simultaneously improving application performance. This is a significant breakthrough, as reducing cloud costs and enhancing performance are often seen as mutually exclusive goals.

StormForge’s Expanding Ecosystem

Beyond bi-directional scaling, StormForge is also strengthening its presence within the cloud ecosystem. The company recently partnered with Amazon Web Services (AWS) and Datadog, making its solution readily available through the AWS Marketplace and Datadog Marketplace, respectively.

  • The AWS partnership allows users of Amazon’s Elastic Kubernetes Service (EKS) to easily integrate StormForge’s optimization capabilities.
  • The Datadog integration leverages the rich observability data collected by Datadog to power StormForge’s real-time optimization recommendations.

A Look into the Future: Autonomous Optimization

Bergstrom shared that StormForge is actively working with a client to deploy its Optimize Live product, which allows for autonomous optimization in production environments. This means that StormForge’s system can automatically analyze, recommend, and execute scaling adjustments without human intervention. This is a significant step toward a fully automated and optimized Kubernetes infrastructure.

Conclusion

StormForge’s bi-directional scaling feature and its growing partnerships represent a significant leap forward in Kubernetes optimization. By enabling the seamless integration of vertical and horizontal scaling, StormForge empowers organizations to maximize the efficiency, performance, and cost-effectiveness of their Kubernetes deployments. The future looks bright for StormForge as it continues to innovate and expand its capabilities, pushing the boundaries of what’s possible in the world of cloud-native applications.

Check out this other episode about scaling Kubernetes.

🔥 Like and Subscribe 🔥

Connect with me 👋
TWITTER ► https://bit.ly/3HmWF8d
LINKEDIN COMPANY ► https://bit.ly/3kICS9g
LINKEDIN PROFILE ► https://bit.ly/30Eshp7

Want to support the show? Buy Me A Coffee! https://bit.ly/3NadcPK

🔗 Links: