Collecting resource metrics

So far, we have seen that serverless and our autonomous service patterns provide us with a lot of fine-grained metrics without the need for explicit instrumentation. Now we need to turn all this data into actionable information so that we can fail forward fast.Back in 2015, when I first found myself in the role of operation manager, I went looking for guidance so that I could put together a mental model for reasoning about all this data. One of the best sources I found was a series of posts by Alexis Lê-Quôc, Datadog's CTO and co-founder, called Monitoring 101 (https://www.datadoghq.com/blog/monitoring-101-collecting-data).In this series he recommends grouping observability data into three main categories: work metrics, resource metrics, and system events. In this section we will dig in resource metrics and cover the USE method and how it applies to serverless and our autonomous service patterns. Specifically, we will look at capacity and concurrency...