Write a Blog >>
Tue 20 Jun 2017 14:25 - 14:50 at Actes, Civil Engineering - Parallelization and Concurrency Chair(s): Milind Kulkarni

We give a performance analysis of the purely functional ar- ray language Futhark and its GPU-targeting optimising com- piler on 16 benchmarks, and present several features that enable performance comparable with hand-written code: (i) a simple type system for in-place updates that ensures referential transparency and supports equational reasoning, (ii) several bulk-parallel operators, which encode strength- reduction invariants, along with their fusion rules, and (iii) a flattening transformation aimed at enhancing the degree of parallelism, which builds on loop interchange and dis- tribution but uses higher-order reasoning rather than array- dependence analysis, and preserves the opportunities for fur- ther locality-of-reference optimizations.

Tue 20 Jun

Displayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change

14:00 - 15:40
Parallelization and ConcurrencyPLDI Research Papers at Actes, Civil Engineering
Chair(s): Milind Kulkarni Purdue University
14:00
25m
Talk
Synthesis of Divide and Conquer Parallelism for Loops
PLDI Research Papers
Azadeh Farzan University of Toronto, Victor Nicolet University of Toronto
Media Attached
14:25
25m
Talk
Futhark: Purely Functional GPU-programming with Nested Parallelism and In-place Array Updates
PLDI Research Papers
Troels Henriksen DIKU, University of Copenhagen, Niels G. W. Serup DIKU, University of Copenhagen, Martin Elsman Department of Computer Science, University of Copenhagen, Fritz Henglein DIKU, Denmark, Cosmin Oancea DIKU, University of Copenhagen
Media Attached
14:50
25m
Talk
Gradual Synthesis for Static Parallelization
PLDI Research Papers
Grigory Fedyukovich UW CSE, Maaz Bin Safeer Ahmad UW / CSE, Rastislav Bodik University of Washington
Pre-print Media Attached
15:15
25m
Talk
Verifying invariants of lock-free data structures with rely-guarantee and refinement type
PLDI Research Papers
Colin Gordon Drexel University, Michael D. Ernst University of Washington, USA, Dan Grossman University of Washington, Matthew Parkinson Microsoft Research, UK
Pre-print Media Attached