You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Do you mean, "where the network is involved"? To what end? This is an
important question, because a remote benchmark on one network won't tell
you much at all about what you can expect on a different network with
different characteristics. What do you expect from transport generic remote
communication benchmarks that local benchmarks don't already tell us with
stronger effects?
As a user, I want to use network-transport-benchmarks as a base for a different benchmarks for n-t implementation starting from the basic benchmarks that measure properties of the transport itself finishing with more complicated tests that benchmark network communication also.
Currently test is organized in a way that it's possible to use cross transport communication (it's up to user setup), but it's not possible to use transport on different hosts. And I think that it will be nice if n-t-benchmark will provide basic building tools for such tests, like latency/throughput tests. So once I'm testing a new network, required property for n-t-implementation I will not reinvent a wheel.
We need to have benchmarks where we could gather statistics for the case where network communication evolved.
The text was updated successfully, but these errors were encountered: