The 5th ACM SIGPLAN Workshop on Functional High-Performance Computing workshop aims at bringing together researchers exploring uses of functional (or more generally, declarative or high-level) programming technology in application domains where high performance is essential. The aim of the meeting is to enable sharing of results, experiences, and novel ideas about how high-level, declarative specifications of computationally challenging problems can serve as maintainable and portable code that approaches (or even exceeds) the performance of machine-oriented imperative implementations.
All aspects of performance critical programming and parallel programming are in-scope for the workshop, irrespective of hardware target. This includes both traditional large-scale scientific computing (HPC), as well as work targeting single node systems with SMPs, GPUs, FPGAs, or embedded processors. FHPC 2016 seeks to encourage a range of submissions, focussing on work in progress and facilitating early exchange of ideas and open discussion on innovative and/or emerging results. Experience reports are also welcome.
More information at the external website: FHPC 2016
Conference DayThu 22 SepDisplayed time zone: Osaka, Sapporo, Tokyo change
09:15 - 10:15 | |||
09:15 60mTalk | From Identification of Parallelizability to Derivation of Parallelizable Codes FHPC Akimasa MorihataUniversity of Tokyo, Japan | ||
10:35 - 11:25 | Domain Specific LanguagesFHPC at Conference Room 2 Chair(s): Zhenjiang HuNational Institute of Informatics, Japan | ||
10:35 25mTalk | Icicle: Write Once, Run Once FHPC | ||
11:00 25mTalk | Using Fusion to Enable Late Design Decisions for Pipelined Computations FHPC | ||
11:45 - 12:35 | |||
11:45 25mTalk | Automatic Generation of Efficient Codes from Mathematical Descriptions of Stencil Computation FHPC Takayuki MuranushiRIKEN, Seiya NishizawaRIKEN, Hirofumi TomitaRIKEN, Keigo NitadoriRIKEN, Masaki IwasawaRIKEN, Yutaka Maruyama, Hisashi YashiroRIKEN, Yoshifumi NakamuraRIKEN, Hideyuki HottaUniversity of Chile, Chile, Junichiro MakinoKobe University, Natsuki HosonoKyoto University, Hikaru InoueFujitsu Limited | ||
12:10 25mTalk | JIT Costing Adaptive Skeletons for Performance Portability FHPC Patrick MaierUniversity of Glasgow, John Magnus MotronUniversity of Glasgow, Phil TrinderUniversity of Glasgow | ||
14:00 - 14:50 | |||
14:00 25mTalk | Low-Level Functional GPU Programming for Parallel Algorithms FHPC Martin Dybdal, Martin ElsmanDepartment of Computer Science, University of Copenhagen, Bo Joel SvenssonChalmers University of Technology, Sweden, Mary SheeranChalmers University of Technology | ||
14:25 25mTalk | APL on GPUs: A TAIL from the Past, Scribbled in Futhark FHPC Troels HenriksenDIKU, University of Copenhagen, Martin Dybdal, Henrik UrmsDIKU, University of Copenhagen, Anna Sofie KiehnDIKU, University of Copenhagen, Daniel GavinDIKU, University of Copenhagen, Hjalte AbelskovDIKU, University of Copenhagen, Martin ElsmanDepartment of Computer Science, University of Copenhagen, Cosmin OanceaDIKU, University of Copenhagen | ||
15:20 - 16:10 | |||
15:20 25mTalk | Streaming Nested Data Parallelism on Multicores FHPC | ||
15:45 25mTalk | Polarized Data Parallel Data Flow FHPC Ben LippmeierUniversity of New South Wales, Fil MackayVertigo Technology (Australia), Amos RobinsonAmbiata (Australia) | ||
16:40 - 17:05 | |||
16:40 25mTalk | s6raph: Vertex-Centric Graph Processing Framework with Functional Interface FHPC Onofre Coll RuizKochi University of Technology, Kiminori MatsuzakiKochi University of Technology, Japan, Shigeyuki SatoKochi University of Technology | ||
17:05 - 18:00 | |||
17:05 55mTalk | Discussion FHPC | ||
Call for Papers
Please see FHPC website
Accepted Papers
Paper Download
FHPC 2016- Proceedings of the 5th International Workshop on Functional High-Performance Computing
Full Citation in the ACM Digital Library
SESSION: Keynote
From identification of parallelizability to derivation of parallelizable codes
SESSION: Domain-Specific Languages
Icicle: write once, run once
Using fusion to enable late design decisions for pipelined computations
SESSION: Code Generation
Automatic generation of efficient codes from mathematical descriptions of stencil computation
JIT costing adaptive skeletons for performance portability
SESSION: GPUs
Low-level functional GPU programming for parallel algorithms
APL on GPUs: a TAIL from the past, scribbled in Futhark
SESSION: Streaming and Dataflow
Streaming nested data parallelism on multicores
Polarized data parallel data flow
SESSION: Graph Processing
s6raph: vertex-centric graph processing framework with functional interface
| Sun 31 Jul 2016 Final copy due |
| Fri 8 Jul 2016 Author notification |
| Fri 10 Jun 2016 Submission Deadline |
David DukeCo-chair
University of Leeds
Yukiyoshi KameyamaCo-chair
University of Tsukuba
Baris Aktemur
Ozyegin University
Marco Aldinucci
University of Torino
Jost Berthold
Commonwealth Bank, Australia
Kei Davis
Los Alamos National Laboratory
Kento Emoto
Kyushu Institute of Technology, Japan
Zhenjiang Hu
National Institute of Informatics, Japan
Japan
Ben Lippmeier
University of New South Wales
Rita Loogen
University of Marburg
Geoffrey Mainland
Drexel University
Mike Rainey
Inria, France
Mary Sheeran
Chalmers University of Technology
Satnam Singh
Facebook