HIPERFITTalk on Nessie by John Reppy
HIPERFIT Seminar Talk
Title: Nessie: A NESL to CUDA Compiler
Presenter: John Reppy, Computation Institute, University of Chicago
Time: Wednesday, December 14th, 2016, 13:15-15:00
Place: Universitetsparken 1, Small Auditorium, University of Copenhagen.
Abstract:
Modern GPUs provide supercomputer-level performance at commodity prices, but they are notoriously hard to program. To address this problem, we have been exploring the use of Nested Data Parallelism (NDP), and specifically Guy Blelloch’s first-order functional language NESL, as a way to raise the level of abstraction for programming GPUs. Preliminary results suggest that NDP can be effectively mapped onto GPUs, but there is significant work required to make this mapping competitive with hand-written GPU code. This talk describes ongoing work on a new compiler for the NESL language that generates CUDA code. Specifically, we will describe several aspects of the compiler that address some of the challenges of generating efficient NDP code for GPUs.
This work is joint with Nora Sandler and Joeseph Wingerter.
Notice: The presentation is part of the course on Parallel Functional Programming (PFP) at DIKU.
Biography
John Reppy is a Professor of Computer Science and a Senior Fellow of the Computation Institute at the University of Chicago. He received his Ph.D. from Cornell University in 1992 and spent the first eleven years of his career at Bell Labs in Murray Hill NJ. He has been exploring issues in language design and implementation since the late 1980’s, with a focus on higher-order, typed, functional languages. His work includes the invention of Concurrent ML and work on combining object-oriented and functional language features. His current research is on high-level languages for parallel programming, including the Manticore and Diderot projects.
Host: Martin Elsman, DIKU, HIPERFIT.