Metrics at a fintech processing billions of dollars of daily GPV, plus the signals from every microservice in the constellation are enormous. Huge scale time series data.

We had an in-house system that worked, but it was a two pizza team split between time series and logging. "Internal weirdware" got thrown around a lot, so we outsourced to SignalFx for a few years. It was bumpy. I liked our in-house system better, and I didn't build it.

Splunk then buys SignalFx and immediately multiplies the pricing at a conveniently timed contract renewal. Suddenly every team in the company has to plan an emergency migration.

What agents are you using? If you stick to opentelemetry and open source agents and develop a collector infrastructure - You can switch across different vendors with lower impact and ramp off time.

Your supply chain is messed up. You need sign longer contracts with price guarantees.