Nesl: A language for teaching parallel algorithms

Guy Blelloch and Jonathan Hardwick, Carnegie-Mellon University

Nesl is a high-level parallel language designed for the teaching and rapid development of parallel algorithms. The language has several features that make it particularly well-suited for describing parallel algorithms, including the ability to use nested data-parallelism to tackle irregular problems, and a formal method of deriving the complexity of a program in terms of work and steps. Nesl has been used to teach graduate and undergraduate-level courses in parallel algorithms for the past three years, and we have developed a library of standard PRAM algorithms using the language, including algorithms for sorting, graphs, trees, and computations geometry. In this talk I will give an overview of Nesl, its features, and implementation, and will demonstrate some interactive algorithm animations developed in the language.