Charm++ for Productivity and Performance: A Submission to the 2011 HPC Class II Challenge

We present our implementation of the HPC Challenge Class II (productivity) benchmarks in the Charm++ programming paradigm. Our submission focuses on explaining how over-decomposed, message-driven, migratable objects enhance the clarity of expression of parallel programs and also enable the runtime system to deliver portable performance. Our submission includes implementations of three required benchmarks: Dense LU Factorization, FFT, and Random Access. We also include two additional benchmarks that represent relevant scientific computing algorithms of some complexity: Molecular Dynamics and Barnes-Hut. We believe our implementations demonstrate that a high-level productivity oriented model can also deliver portable performance via an intelligent runtime.