2 ms·
You could gather and periodically log a latency histogram for each type of request: https://www.fsl.cs.sunysb.edu/project-osprof.html https://www.fsl.cs.sunysb
by hedora 5y ago
You could gather and periodically log a latency histogram for each type of request:
https://www.fsl.cs.sunysb.edu/project-osprof.html https://www.fsl.cs.sunysb.edu/project-osprof.html
This covers all the examples you gave except CDN latency. You could try polling CDN endpoints from an external client that also gathers histograms.