Big-step and Small-Step Semantics of the Call-by-Name RPC Calculus

Abstract

A remote procedure call (RPC) is a network communication technique between distributed computers. RPC is more approachable than the other network communication techniques since a programmer can use it in a similar manner to a procedure call in a sequential program on a single CPU computer. Cooper and Wadler proposed the RPC calculus and formalized the remote procedure call in the style of the lambda calculus. They used the call-by-value evaluation strategy for the RPC calculus. We may say that the RPC calculus is an extension of the traditional call-by-value lambda calculus by attaching a location. In the previous work, we developed a big-step semantics of the call-by-name RPC calculus and studied the translation of the call-by-name RPC calculus into the call-by name RMI calculus, in order to show the expressive power of the RMI calculus. In this paper, we newly propose a small-step semantics of the call-by-name RPC calculus. We prove the equivalence between the small-step and big-step semantics.