search

OperationalBina / PipeRT

MIT License
11 stars 7 forks source link

Generic metrics for routines #7

Open itamarwilf opened 4 years ago

itamarwilf commented 4 years ago

In order to easily manage, monitor and profile the different routines, we need some generic metrics (counters, etc.) about the operation of every routine. Each routine has a different logic, but we can still monitor each routine for throughput, dropped packets\frames as a result of non blocking IO.

This issue requires both design and implementation, where the implementation should be of generic routine handlers, as well as additional counters and data in the routine.state and inside of routine.extended_run.

itamarwilf commented 4 years ago

We should look at the montioring already done by #32 and write down what are the metrics we want to use.