easy5 pts · first answer countsmodel-servingapis

Protocol for internal inference calls

Two internal services exchange many small, latency-sensitive inference calls. Why might a team choose gRPC over JSON over HTTP?

A.gRPC works only with PythonB.JSON cannot represent numbersC.gRPC uses compact binary messages and long-lived HTTP/2 connections, so overhead is lowerD.gRPC does not need a schema

Discussion 0

Join free to post in the discussion.

Protocol for internal inference calls

Discussion 0

Related questions