FHPC 2013

Background

The FHPC workshop aims at bringing together researchers exploring uses of functional (or more generally, declarative or high-level) programming technology in application domains where large-scale computations arise naturally and high performance is essential. Such computations would typically -- but not necessarily -- involve execution on highly parallel systems ranging from multi-core multi-processor systems to graphics accelerators (GPGPUs), reconfigurable hardware (FPGAs), large-scale compute clusters or any combination thereof. It is becoming apparent that radically new and well founded methodologies for programming such systems are required to address their inherent complexity and to reconcile execution performance with programming productivity.

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 highly transparent, maintainable, and portable code that approaches (or even exceeds) the performance of machine-oriented imperative implementations.

Each FHPC workshop proposes a particular theme for applications where high-performance computing and/or functional programming technology can be applied. For FHPC 2013, the theme is "Large-Scale Simulation", traditionally one of the main driving forces behind supercomputing. A large fraction of compute cycles in supercomputers worldwide is spent on simulation tasks, for various engineering tasks, drug design and other medical simulations, and in different natural science domains. Declarative languages have potential to radically change development practice and workflow for simulation software in these areas. Hence, we particularly encourage submission of application-oriented contributions in the area of simulation.

As a general rule, while proposing the theme, the workshop welcomes submissions from all relevant application domains as well as those describing general work on the theory and practice of declarative high-performance computing.

FHPC 2013 is the second workshop in the FHPC series. The first workshop was run in 2012, affiliated with ICFP in Copenhagen.

Submission and publication

Contributions should be submitted via Easychair. Submitted papers must be in portable document format (PDF), formatted according to the ACM SIGPLAN style guidelines (double column, 9pt format). See the SIGPLAN Author Information page for more information and style files. The page limit is 12 pages. Any paper submitted must adhere to ACM SIGPLAN's republication policy. Submission deadlines and page limit are firm.

Accepted papers will be published by the ACM and will appear in the ACM Digital Library.

The deadline has passed, the FHPC'13 submission site is closed.

Important dates

• Submission deadline: 19 June 2013 (anywhere on Earth), anywhere on Earth.
• Author notification: 11 July 2013
• Final papers due: 28 July 2013 (instructions for authors)
• Workshop: 23 September 2013

Travel Support

Student attendees with accepted papers can apply for a SIGPLAN PAC grant to help cover travel expenses. PAC also offers other support, such as for child-care expenses during the meeting or for travel costs for companions of SIGPLAN members with physical disabilities, as well as for travel from locations outside of North America and Europe. For details on the PAC programme, see http://www.sigplan.org/PAC.htm.

Workshop programme

9:00 - 10:30 — Runtime Techniques for Declarative Parallel Programming

Matthew Fluet.

The Manticore Project(invited talk)

Sylvain Henry.

ViperVM: a Runtime System for Parallel Functional High-Performance Computing on Heterogeneous Architectures

11:00 - 12:30 — Parallel Programming Models and Application Classes

Frederik M. Madsen and Andrzej Filinski.

Towards a Streaming Model for Nested Data Parallelism

Qi Wang, Meixian Chen, Yu Liu and Zhenjiang Hu.

Towards Systematic Parallel Programming of Graph Problems via Tree Decomposition and Tree Parallelism

Josef Svenningsson, Joel Svensson and Mary Sheeran.

Counting and Occurrence sort for GPUs using an Embedded Language

13:30 - 14:30 — Optimizing Compilation of Functional Programs

Troels Henriksen and Cosmin E. Oancea.

A T2 Graph-Reduction Approach To Fusion

Artjoms Sinkarovs and Sven-Bodo Scholz.

Sematics-Preserving Data Layout Transformations for Improved Vectorisation

14:30 - 15:30 — Libraries for Parallel Functional Programming

Lindsey Kuper and Ryan R. Newton.

LVars: Lattice-based Data Structures for Deterministic Parallelism

Mauro Blanco, Pablo Perdomo, Pablo Ezzatti, Alberto Pardo and Marcos Viera.

Towards a functional run-time for dense NLA domain

16:00 - 17:30 — Data Parallelism

Manuel Chakravarty.

Data Parallelism in Haskell(invited talk)

Panel: (Clemens Grelck -UvA, Manuel Chakravarty -UNSW, Neal GlewR.Newton -Ind.U, R.S.Nikhil -Bluespec, Cosmin Oancea -DIKU)

Data Parallelism and GPU Computing

A detailed programme with abstracts is available on a separate page.

Workshop organisation

Programme committee

• Umut Acar (co-chair), Carnegie Mellon University, PA, USA (umut@cs.cmu.edu)
• Arvind, MIT, MA, USA
• Jost Berthold (co-chair), U. of Copenhagen, Denmark (berthold@diku.dk)
• Guy Blelloch, Carnegie Mellon U., PA, USA
• Hassan Chafi, Oracle Labs, CA, USA
• Dan Spoonhower, Google, CA, USA
• Sergei Gorlatch, U. Münster, Germany
• Clemens Grelck, U. of Amsterdam, Netherlands
• Vinod Grover, NVidia, USA
• Torsten Grust, U.Tübingen, Germany
• Zhenjiang Hu, National Inst. of Informatics, Tokyo, Japan
• Suresh Jagannathan, Purdue U., USA
• Gabriele Keller, U. New South Wales, Sydney, Australia
• Yaron Minsky, Jane Street Capital, NY, USA
• Jens Palsberg, U. of California, CA, USA
• Leaf Peterson, Intel, USA
• Mike Rainey, MPI-SWS,Kaiserslautern, Germany
• Sven-Bodo Scholz, Heriot-Watt U., Edinburgh, UK
• Guy Steele, Oracle Labs, Burlington, MA, USA

General Chairs

• Fritz Henglein, Dept. of Computer Science (DIKU), University of Copenhagen, Denmark; henglein@diku.dk
• Clemens Grelck, Informatics Instititute, University of Amsterdam, Netherlands; c.grelck@uva.nl