AI and Performance Engineering: How can AI gather enough data to help make informed decisions?

Data. Today there is more of it than we could possibly process, and yet SREs are still struggling to make informed decisions about how to improve the performance and efficiency of Kubernetes applications. Oh, and don’t forget optimizing costs! Because with all the data available, organizations that adopt Kubernetes are still struggling to manage cloud costs that are spiraling out of control. 

What we have isn’t a data problem. It’s a lack of ways to gather, analyze, and make informed decisions based upon more data than ever. The challenge is that cloud native and DevOps adds a layer of complexity to the areas SREs are charged with analyzing – performance testing and performance engineering. The advent of microservices has only served to take the problem and multiply it many times over. There are more variables to analyze and, as humans, we aren’t equipped to process the vast number of variables cloud native workloads entail. 

Fortunately, with machine learning, we don’t have to. Machine learning and automation accelerate the collection and analysis of data to allow SREs to apply business goals to application and cloud investments. 

Every Kubernetes application is unique and complex, which is why imagining all of the configuration variables you can apply to it can be mind-blowing and overwhelming. ML and automation change the game with the ability to quickly, efficiently and cost-effectively identify the variables, recognize the dependencies, and ascertain the impact of decisions moving forward and help with performance/cost trade-offs. 

Data is an SRE’s friend. Machine learning and automation turn data into actionable intelligence. And that’s the difference between collecting data and making better business decisions. 

Watch the video from Scott Moore’s Performance Tour, where Patrick Tavares discussed how–and how AI can gather the data needed to make informed decisions – and how StormForge is leading the way.