MULTILISP: a language for concurrent symbolic computation

Abstract

Multilisp is a version of the Lisp dialect Scheme extended with constructs for parallel execution. Like Scheme, Multilisp is oriented toward symbolic computation. Unlike some parallel programming languages, Multilisp incorporates constructs for causing side effects and for explicitly introducing parallelism. The potential complexity of dealing with side effects in a parallel context is mitigated by the nature of the parallelism constructs and by support for abstract data types: a recommended Multilisp programming style is presented which, if followed, should lead to highly parallel, easily understandable programs.Multilisp is being implemented on the 32-processor Concert multiprocessor; however, it is ultimately intended for use on larger multiprocessors. The current implementation, called Concert Multilisp, is complete enough to run the Multilisp compiler itself and has been run on Concert prototypes including up to eight processors. Concert Multilisp uses novel techniques for task scheduling and garbage collection. The task scheduler helps control excessive resource utilization by means of an unfair scheduling policy; the garbage collector uses a multiprocessor algorithm based on the incremental garbage collector of Baker.

Controlling Concurrency and Expressing Synchronization in Charm++ Programs

“…The future construct was described, in the sense we use it, in the multiLisp system of Halstead [15], although multiple precursors existed before that. Taura, Matsuoka, and Yonezawa [46] extended ABCL to support the future construct as well.…”

Implementing Joins Using Extensible Pattern Matching

“…Get()) // synchronous send, waits for reply Synchronous message sends make the sending process wait for the actor to reply to the message (by means of reply(value)). Scala actors also offer more advanced synchronization patterns such as futures [12,25]. actor !!…”

Design and Performance of Asynchronous Method Handling for CORBA

“…An initial inspiration for this approach was the implementation of continuations in the MACH kernel, which resulted in significant performance improvements in that OS kernel [5]. Other examples of similar work include Futures [18] and Promises [19], which are language mechanisms that decouple method invocation from method return values passed back to the caller when a method finishes executing.…”