Write a Blog >>
ICFP 2016
Sun 18 - Sat 24 September 2016 Nara, Japan
Thu 22 Sep 2016 15:20 - 15:45 at Conference Room 2 - Streaming and Dataflow Chair(s): Hai Liu

The paradigm of nested data parallelism (NDP) allows a variety of semi-regular computation tasks to be mapped onto SIMD-style hardware, including GPUs and vector units. However, some care is needed to keep down space consumption in situations where the available parallelism may vastly exceed the available computation resources. To allow for an accurate space-cost model in such cases, we have previously proposed the Streaming NESL language, a refinement of NESL with a high-level notion of streamable sequences.

In this paper, we report on experience with a prototype implementation of Streaming NESL on a 2-level parallel platform, namely a multicore system in which we also aggressively utilize vector instructions on each core. We show that for several examples of simple, but not trivially parallelizable, text-processing tasks, we obtain single-core performance on par with off-the-shelf GNU Coreutils code, and near-linear speedups for multiple cores.

Thu 22 Sep

Displayed time zone: Osaka, Sapporo, Tokyo change

15:20 - 16:10
Streaming and DataflowFHPC at Conference Room 2
Chair(s): Hai Liu Intel Labs
Streaming Nested Data Parallelism on Multicores
Frederik M. Madsen DIKU, University of Copenhagen, Andrzej Filinski DIKU, University of Copenhagen
Polarized Data Parallel Data Flow
Ben Lippmeier University of New South Wales, Fil Mackay Vertigo Technology (Australia), Amos Robinson Ambiata (Australia)