Parallelism in LISP

1. Parallelism in Lisp There are two main approaches to executing Lisp in parallel. One is to use existing code and clever compiling methods to parallelize the execution of the code [9, 14, 11]. This approach is very attractive because it allows the use of already existing code without modification, and it relieves the programmer from the significant conceptual overhead of parallelism. This approach, known as the "dusty deck" approach, suffers from a simple problem: it is very hard to do. This is particularly true in a language such as Lisp that shows a much less well defined flow of control than languages such as FORTRAN where such techniques have been applied relatively successfully [13]. A result of this is that, given the current compiler techniques, the amount of parallelism that can be achieved is limited.