While you can learn a lot by emitting metrics from your application, some insights can only be gained by looking at OS metrics. Yet OS metrics, despite being commonly used are also frequently misunderstood.
In this hands-on workshop, we go over commonly used CPU, memory, disk and network metrics, and make sure we understand each of them. We experiment with old & new tools to acquire and analyse metrics, evaluate a black-box workloads by looking at metrics, and assess the effect of extreme metric values on simple applications. During our adventure, we learn about Linux internals, it’s underlying optimizations and unexpected limitations.
(Co-Authored with Nati Cohen (@nocoot))