There are many great tools for logs and metrics in the K8s ecosystem, but when there’s a problem, skilled SREs and developers still spend a lot of time searching through logs and dashboards to find root cause. The underlying challenge is the near infinite number of possible failure modes in distributed applications.
Now, imagine a machine learning system that could uncover root cause just by watching a feed of logs and metrics? And imagine if it didn’t require any training or setup? In this webinar, we will discuss a number of machine learning techniques for logs and metrics and then demonstrate what this looks like in real life. Spoiler alert: ML really can detect problems and their root cause unaided!
Larry Lancaster, Founder and CTO @Zebrium