Title: HLPGPU 2019: High-Level Programming for Heterogeneous and Hierarchical Parallel Systems
AIMS AND SCOPE
HLPGPU aims to bring together researchers and practitioners to present new results and ongoing
work on those aspects of high-level programming relevant, or specific, to
GPGPUs and new architectures.
- High-level parallel programming models, libraries and languages.
- Adaptation to heterogeneous/hierarchical platforms: GPGPU, multicore,
FPGA etc.
- Timing and energy models for heterogeneous multi cores
- Type systems for energy and/or timing models
- Adaptation of cluster- and grid- performance models to new platforms.
- Programming experiments and applications of heterogeneous/hierarchical
platforms with a view on high-level programming methods.
- Algorithmic skeletons, design patterns for heterogeneous/hierarchical
platforms.
- Domain-specific parallel patterns
- Efficient implementations of parallel patterns on multicore systems
- Mapping patterns to heterogeneous multicore systems
- Exploiting performance information to guide pattern Implementation
- Pattern implementations using application-specific information
- Refactoring tools and techniques for parallel patterns
- Tools and systems to aid pattern-based parallel programming - structured
parallel programming models
- Heterogeneous computing
- Feedback-directed compilation
- Timing analysis tools and techniques
- Statistical performance modelling and prediction
PROGRAMME
10:00 Marco Aldinucci (University of Torino) Keynote talk.
11:00 - 11:30 Coffee Break.
11:30 - Tim Besard and Bjorn De Sutter. High-level language design for extensible accelerator programming.
12:00 - Andrés Rodriguez, Angeles Navarro, Rafael Asenjo, Francisco Corbera, Ruben Gran, Darío Suárez Gracia and Jose Nunez-Yanez. Evaluation of Heterogeneous execution on an HPC-oriented CPU-FPGA System-on-Chip.
12:30 - Demetrios Coutinho, Kris Nikov, Kyriakos Georgiou, Jose L. Nunez-Yanez, Kerstin I. Eder and Samuel Xavier-de-Souza. Analytic Models for Exposing Performance and Energy Efficiency Trade-Offs in Heterogeneous Multi-Processing.
13:00 - 14:00 Lunch Break.
14:00 - Marco Danelutto, Horacio Gonzalez-Velez, Gabriele Mencagli, Peter Kilpatrick and Massimo Torquati. Algorithmic skeletons & Parallel design patterns in everyday parallel programming.
14:30 - Vladimir Janjic, Christopher Brown, Adam Barwell and Kevin Hammond. Software Restoration of Legacy-Parallel Pthreads in Parallel Applications.
15:00 - Nikos Fragkoulis, Vasileios Pothos, Ilias Theodorakopoulos and Evangelos Vassalos. Inference on Edge with Dynamic Graphs. Can it be Beneficial?
15:30 - 16:00 Coffee Break.
16:00 - Adam Seewald, Emad Ebeid and Ulrik Schultz. Dynamic Energy Modelling for SoC Boards: Initial Experiments.
16:30 Close.
SUBMISSION
Submission is now Closed.
We will reopen for a special issue shortly.
IMPORTANT DATES
- Workshop 23rd January, 2019
ORGANISERS
Chris Brown (cmb21@st-andrews.ac.uk) (University of St Andrews)
Adam Barwell (adb23@st-andrews.ac.uk) (University of St Andrews)
PROGRAMME COMMITTEE
- Chris Brown (University of St Andrews) (PC Chair) -
- J. Daniel Garcia (University Carlos III Madrid)
- Murray Cole (University of Edinburgh)
- Heiko Falk (TUHH, Germany)
- Clemens Grelck (Universiteit van Amsterdam) -
- Olivier Zendra (INRIA)
- Kevin Hammond (University of St Andrews)
- Herbert Kuchen (University Muenster)
- Arturo González Escribano (Unversity of Valladolid)
- Marco Aldinucci (University of Pisa)
- Ulrik Schultz (University of Southern Denmark)
- Michael Rossbory (SCCH)