Sun 4 Nov 2018 10:55 - 11:20 at Stuart - I Chair(s): Mark Marron

Parallel skeletons are essential structured design patterns for efficient heterogeneous and parallel programming. They allow programmers to express common algorithms in such a way that it is much easier to read, maintain, debug and implement for different parallel programming models and parallel architectures. Reductions are one of the most common parallel skeletons. Many programming frameworks have been proposed for accelerating reduction operations on heterogeneous hardware. However, for the Java programming language, little work has been done for automatically compiling and exploiting reductions in Java applications on GPUs.

In this paper we present our work in progress in utilizing compiler snippets to express parallelism on heterogeneous hardware. In detail, we demonstrate the usage of Graal’s snippets, in the context of the Tornado compiler, to express a set of Java reduction operations for GPU acceleration. The snippets are expressed in pure Java with OpenCL semantics, simplifying the JIT compiler optimizations and code generation. We showcase that with our technique we are able to execute a predefined set of reductions on GPUs within 85% of the performance of the native code and reach up to 20x over the Java sequential execution.

Sun 4 Nov

Displayed time zone: Guadalajara, Mexico City, Monterrey change

10:30 - 12:00
IVMIL at Stuart
Chair(s): Mark Marron Microsoft Research
10:30
25m
Research paper
Efficient VM-independent Runtime Checks for Parallel Programming
VMIL
Michael Faes ETH Zurich, Thomas Gross ETH Zurich
DOI Pre-print
10:55
25m
Research paper
Using Compiler Snippets to Exploit Parallelism on Heterogeneous Hardware: A Java Reduction Case Study
VMIL
Juan Fumero The University of Manchester, Christos Kotselidis The University of Manchester
DOI Pre-print
11:20
20m
Talk
Generating a Minimum JavaScript VM Specialised for Target Applications
VMIL
Tomoharu Ugawa Kochi University of Technology, Japan, Hideya Iwasaki University of Electro-Communications, Japan
11:40
20m
Talk
Profiling Android Applications with Nanoscope
VMIL
Lun Liu University of California at Los Angeles, USA, Leland Takamine Uber Technologies, Adam Welc Uber Technologies
Pre-print