Deep Dive into Machine Learning Kubernetes Optimization at Cloud Field Day 13

StormForge recently had the opportunity to present an overview of our platform for optimizing Kubernetes resource management at Cloud Field Day 13. The presentation was well-received by the Cloud Field Day delegates, with lots of great questions and interaction.

In this blog, I’ll summarize what we shared and some of the feedback. Additionally, you can watch each of the presentations embedded below.

Solving the Kubernetes Efficiency Problem with StormForge

As enterprises scale up their cloud native environments for day 2 operations, cloud waste and inefficiency often become a significant problem. The business impacts of inefficiency include cloud costs that are too high, risk of performance and availability issues that affect user experience, and pulling developers away from innovating to spend time on manual tuning of Kubernetes applications. In this fifteen-minute session, I provide an overview of the StormForge platform for Kubernetes optimization, and how it improves efficiency by applying machine learning for both non-production experimentation and scenario analysis as well as production optimization leveraging observability data already being collected.

Using Machine Learning for Kubernetes Optimization: StormForge Product Demo

The StormForge platform uses patent-pending machine learning to automate the process of achieving and maintaining efficiency of cloud native environments running on Kubernetes. As the only solution that combines experimentation-based, non-production optimization with observation-based optimization in production, StormForge helps DevOps teams to incorporate optimization as a systematic, continuous process that closes the loop between production and non-production and turns observability into actionability. In this session, Sr. Solutions Architect Patrick Tavares explains how the StormForge platform works, showing how easy it is to tackle the problems of cloud waste, inefficient resource utilization, performance and availability risk, and time-consuming manual application tuning.

Kubernetes Optimization at Acquia: An Example StormForge Use Case

Acquia is the largest provider of hosting capabilities for Drupal, with more than 4,000 organizations using the company’s solutions for content, community and commerce. To continue meeting customer needs and optimize the performance, scalability and cost efficiency of its core hosting offering, Acquia decided to move to a Kubernetes container-based platform. This major re-platforming effort would affect 2500+ customers and tens of thousands of applications. With this move, Acquia needed the ability to right-size resource capacity to support customer applications, while also scaling as needed based on each application’s consumption requirements. With each customer environment being unique, manual application tuning wasn’t feasible due to the massive amount of time and effort that would be required. In this session, Director of Technical Alliances Erwin Daria shows how Acquia was able to automate the process of achieving and maintaining Kubernetes resource efficiency at scale. With the ability to forecast demand, and then make resource decisions around application configuration based upon the forecast, Acquia can now effectively manage applications and meet their customers’ evolving requirements.

Delegate Feedback on the StormForge Platform

Following the live Cloud Field Day session, several of the delegates in attendance blogged about StormForge. Here are a few of the highlights.

“I liked the StormForge approach as it allows developers to ensure applications are tested and optimised before being put into production. Goal setting also enables applications to be re-tested and tuned after major code updates. Once in production, Optimize Live ensures applications stay optimised, based on the inevitable unpredictability of production versus test scenarios…. This is one start-up to keep watching.”

– Optimising Kubernetes with StormForge by Chris Evans, Architecting IT

“What stood out most to me about the StormForge Platform is that it takes a holistic approach to operational optimization on Kubernetes. By bringing with it the power of Machine Learning, it causes a well-rounded development of cloud native environment. In the process it saves enterprises from painfully big cloud bills and resource wastage, while accelerating application performance.”

– Optimizing Resources in Kubernetes with StormForge Platform by Sulagna Saha, Gestalt IT

“Less and less people find less and less time to manually finetune complex systems like K8s. That’s where @StormForgeIO comes in with some machine learning things. Looks promising.”

– Wolfgang Stief, Data Disrupted

“The mix of #MachineLearning and actual experimentation tests is really cool. Don’t rely on just models or random tests.”

– Brian Knudtson, iland Cloud