Monitoring user workstations with Grafana Cloud and Prometheus

When we talk about monitoring, we usually picture servers and applications. We actively try to keep infrastructure out of reach, make it more and more immutable, increase automation, and decrease human interaction. But what if you wanted data and insights into… user workstations? What if production business functions with tight SLAs were tied to workstation performance? What if those workstations were sometimes remote? What if you wanted to know how much RAM Chrome really uses?

At Ocrolus, a fintech automation platform, we set out to capture meaningful data from our nearly 1,000 operational workforce users. We wanted to reuse as much existing industry tooling and knowledge as possible (namely Grafana Cloud and Prometheus), and we had to deal with some unique resource constraints, as well as the need to sometimes operate outside any controlled network (e.g., users working from home). In the process, we implemented our first remote configuration of these systems, as well as access and auditing systems.

In this session, we’ll go over how we figured out how to deal with remote networking, and share what meaningful data we were able to collect from these machines.