Haskell as an Introduction to Parallel Computing for Undergraduates

摘要

As computing manufacturers increasingly rely on the use of multiple cores instead of improved clock speeds, programming in parallel has become an essential skill for computer scientists. Insufficient effort has been made to introduce the concept of parallel computing to the undergraduate curricula. We explored the pedagogical potential of a functional programming language, Haskell, in the context of parallel computing education. We extended previous work by providing both sequential and parallel Haskell implementations of a program that uses Riemann integration to calculate π. Our assessment is that the parallel tools of Haskell are both concise and scalable. However, they can be difficult to learn, which brings us to our one reservation about using Haskell as a parallel programming language at an undergraduate level: learning functional programming may distract students from the original focus of learning general parallel programming concepts. As such, we expect that a module on parallel Haskell programming would be more effective if it was introduced in a programming languages course, which would presumably have already introduced students to the basic concepts of functional programming.