In distributed computing, a few layers of abstraction up, an analogous technique of sending two identical RPCs to distinct backends can be used to reduce tail l

In distributed computing, a few layers of abstraction up, an analogous technique of sending two identical RPCs to distinct backends can be used to reduce tail latency.