Want to learn case study under large traffic services

[kanata2]

I would like to lean large scale case study for OpenTelemetry Collector(collector).

I am concerned that even if sampling and other processing is done in the collector layer,
the collector needs to receive all the spans once, which would make the running cost of the collector very expensive.

In fact, we are trying to reduce the running cost of the collector by setting the sampling rate sufficiently small in Istio's MeshConfig and using head-based sampling,
but we are not able to fully utilize the collector's functionality and are looking for a better way to do so.
We are wondering if there is a better way.