HLPP 2014
7th International Symposium on High-level Parallel Programming and Applications
Amsterdam, Netherlands, July 3-4, 2014

Amsterdam


New: Slides of most talks available online

Closed: Registration

Please, visit the HLPP 2014 registration page.

Registation includes:
  • access to all technical sessions
  • keynote
  • coffee breaks on both days
  • lunch on both days
  • social event Thursday afternoon
  • conference banquet dinner  Thursday evening
  • draft proceedings of HLPP 2014
  • for authors: publication in International Journal of Parallel Programming
Should you need a formal invitation letter for visa purposes, please contact the organiser after successful registration.

Keynote:

Speaker:

Abstract: 
The major goal of CnC (Concurrent Collections) is a productive path to efficient parallel execution. And yet a CnC program does not indicate what runs in parallel. Instead, it explicitly identifies what precludes parallel execution. There are exactly two reasons that computations cannot execute in parallel. If one computation produces data that the other one consumes, the producer must execute before the consumer. If one computation determines if another will execute, the controller must execute before the controllee. CnC is a data and control flow model together with tuple-space influence. However, it is closer in philosophy to the PDG (Program Dependence Graph) intermediate form than to other parallel programming models. Its high-level abstractions allow flexible and efficient mapping of a CnC program to the target platform. The talk will introduce CnC and present some active research topics.

Short biography: 
Dr Frank Schlimbach obtained his PhD from University of Greenwich, United Kingdom, in 2000. He started his professional career as a Senior Software Engineer at Pallas GmbH in Germany. Since 2003 Frank is a Software Architect at Intel.

Programme:

 Thursday   July 3, 2014
 09:00 - 09:20  Registration
 09:20 - 09:30  Welcome and Opening 
 09:30 - 10:30  Keynote:
 Frank Schlimbach, Intel:
 Parallelism Through CnC (Concurrent Collections) - More Flexibility, Less Pain

 10:30 - 11:00  Coffee break
 11:00 - 12:30  Session 1: (chair: Quentin Miller)
 Usman Dastgeer and Christoph Kessler: 
 Smart containers and skeleton programming for GPU-based systems 
 Kiminori Matsuzaki and Reina Miyazaki:
 Parallel Tree Accumulations on MapReduce
 Shigeyuki Sato and Kiminori Matsuzaki:
 A Generic Implementation of Tree Skeletons
 12:30 - 14:00      Lunch break: Restaurant Herengracht
 14:00 - 15:30  Session 2: (chair: Rob Bisseling)
 Konrad Siek and Paweł T. Wojciechowski:
 Atomic RMI: a Distributed Transactional Memory Framework
 Tristan Aubrey-Jones and Bernd Fischer:
 Synthesizing MPI Implementations from Functional Data-Parallel Programs
 Antoine Tran Tan, Joel Falcou, Daniel Etiemble and Hartmut Kaiser:
 Automatic Task-based Code Generation for High Performance Domain Specific Embedded Language
 15:30 - 16:00  Coffee break
 16:00 - 17:00  Session 3: (chair: Kiminori Matsuzaki)
 Carlos Alberto Martínez-Angeles, Inês Dutra, Vítor Santos Costa and Jorge Buenabad-Chavez
 Relational Learning with Datalog and GPUs: Accelerating Rule Coverage
 Tarek Menouer, Mohamed Rezgui, Bertrand Le Cun and Jean-Charles Régin:
 Mixing Static and Dynamic Partitionning to Parallelize a Constraint Programming Solver 
 17:30 - 22:00  Social event: walking dinner through Amsterdam
   
 Friday  July 4, 2014
 09:30 - 10:30  Session 4: (chair: Christoph Kessler)
 Marco Aldinucci, Sonia Campa, Marco Danelutto, Peter Kilpatrick and Massimo Torquati:
 Pool evolution: a domain specific parallel pattern  
 Ye Wang and Zhiyuan Li:  
 GridFOR: A Domain Specific Language for Parallel Grid-based Applications
 10:30 - 11:00      Coffee break
 11:00 - 12:30  Session 5: (chair: Gaétan Hains)   
 Jose M. Andion, Manuel Arenaz, Francois Bodin, Gabriel Rodriguez and Juan Tourino:
 Locality-Aware Automaric Parallelization for GPGPU with OpenHMPP Directives

 Alvaro Estebanez, Diego R. Llanos and Arturo Gonzalez-Escribano:
 New Data Structures to Handle Speculative Parallelization at Runtime
 Miguel Areias and Ricardo Rocha:
 A Lock-Free Hash Trie Design for Concurrent Tabled Logic Programs
 12:30 - 14:00  Lunch break: Restaurant Dante's Kitchen
 14:00 - 15:00  Session 6: (chair: Zhiyuan Li)
 Jean Fortin and Frederic Gava:
 BSP-Why: a tool for deductive verification of BSP algorithms with subgroup synchronization
 Ali Jannesari, Nico Korpowski, Felix Wolf, Jochen Schimmel and Walter F. Tichy:
 Generating Parallel Unit Tests for Correlated Variables
 15:00 - 15:15  Closing and outlook towards the future
 Clemens Grelck:
 HLPP 2014 in numbers
 Massimo Torquati:
 Invitation to HLPP 2015 in Pisa, Italy
 Clemens Grelck (on behalf of Herbert Kuchen):
 Looking forward to HLPP 2016 in Schloss Raesfeld, Germany
 Clemens Grelck:
 Farewell
 15:30 - 16:30  HLPP steering committee meeting


Aims and scope of HLPP:

As processor and system manufacturers increase the amount of both inter- and intra-chip parallelism it becomes crucial to provide the software industry with high-level, clean and efficient tools for parallel programming. Parallel and distributed programming methodologies are currently dominated by low-level techniques such as send/receive message passing, or equivalently unstructured shared memory mechanisms. Higher-level, structured approaches offer many possible advantages and have a key role to play in the scalable exploitation of ubiquitous parallelism. 

Since 2001 the HLPP series of workshops/symposia has been a forum for researchers developing state-of-the-art concepts, tools and applications for high-level parallel programming. The general emphasis is on software quality, programming productivity and high-level performance models. The 7th Symposium on High-Level Parallel Programming and Applications will be held July 3-4 in the historic center of Amsterdam. 

Proceedings:

Accepted papers will be distributed as informal draft proceedings during the symposium. All accepted papers will be published by Springer in a special issue of the International Journal of Parallel Programming (IJPP).

Important dates:

  • Submission deadline: April 4 (extended to April 14, anywhere on earth)
  • Author notification: May 12 
  • Camera-ready paper due for draft proceedings: June 20
  • Early registration deadline: June 20
  • Symposium: July 3-4 (Thursday/Friday)
  • Camera-ready paper due for IJPP journal publication: Aug 15

Topics: 

HLPP 2014 invites papers on all topics in high-level parallel programming, its tools and applications including, but not limited to, the following aspects:
  • High-­level programming and performance models (BSP, CGM, LogP, MPM, etc.) and tools
  • Declarative parallel programming methodologies
  • Algorithmic skeletons and constructive methods
  • Declarative parallel programming languages and libraries: semantics and implementation
  • Verification of declarative parallel and distributed programs
  • Software synthesis, automatic code generation for parallel programming
  • Model-driven software engineering with parallel programs
  • High-level programming models for heterogeneous/hierarchical platforms
  • High-level parallel methods for large structured and semi-structured datasets
  • Applications of parallel systems using high­-level languages and tools
  • Teaching experience with high­-level tools and methods 

Paper preparation and submission:

Papers submitted to HLPP2014 must describe original research results and must not have been published or simultaneously submitted anywhere else. Manuscripts must be prepared with the Springer IJSS latex macro package using the single column option (\documentclass[smallextended]{svjour3}) and submitted via the EasyChair Conference Management System as one pdf file. The strict page limit for initial submission and camera-ready version is 20 pages in the aforementioned format.

Each paper will receive a minimum of three reviews by members of the international technical programme committee (see below). Papers will be selected based on their originality, relevance, technical clarity and quality of presentation. At least one author of each accepted paper must register for the HLPP 2014 symposium and present the paper. After the symposium you have ample time to revise the paper incorporating potential comments and remarks of your colleagues. We expect the the HLPP 2014 special issue of the International Journal of Parallel Programming to appear online-first by the end of the year and the printed edition in mid-2015.

Symposium venue:

The HLPP symposium will take place in the Doelenzaal of the University Library of the University of Amsterdam right in the historic heart of Amsterdam opposite the famous flower market. This location is only a short walk away from Amsterdam hotspots such as Rembrandtplein and Leidseplein. It is in good walking distance from Amsterdam's central railway station (Amsterdam Centraal), the Royal Palace as well as the famous Museumplein with the recently re-opened Rijksmuseum, the Stedelijk Museum and the van Gogh-Museum.



Travel to Amsterdam:

Amsterdam's airport Schiphol is the 4th-largest air traffic hub in Europe. So, flying into Amsterdam should be easy wherever you are. Schiphol airport has a railway station right underneath the terminal building. Trains to Amsterdam Centraal depart every few minutes and take 15-20 minutes for less than 4 EUR. Taking a taxi is usually not a good idea: taxis are very expensive and much slower than trains. Taking a taxi from Amsterdam Centraal station to your hotel may be an option, still expensive compared with other countries though.

High speed rail services connect Amsterdam to Brussels (1h50), Paris (3h15), Cologne (2h30) and Frankfurt (4h00). Direct trains from Hannover and Berlin, sleeper services from Copenhagen, Berlin, Warsaw, Moscow, Prague, Munich and Zurich or the overnight ferry from Newcastle can be further alternatives to air travel. See Thalys for connections from Belgium and France, Deutsche Bahn for connections from Germany and the rest of Europe or Nederlandse Spoorwegen for domestic rail services.

Getting around in Amsterdam:

With about 800,000 inhabitants Amsterdam is by far the largest city of the Netherlands, but certainly one of the smaller capital cities in the world. Large parts of the historic center date back to the 17th and 18th century making Amsterdam an open air museum inviting you to stroll alongside the famous canals. Unless you have mobility issues, almost any distance in the historic center of Amsterdam, which can easily be identified on a map as anything inside the outermost canal ring, may be considered walking distance.

Notwithstanding, a large number of tram lines as well as a few bus lines criss-cross central Amsterdam with Amsterdam Centraal being the main hub of exchange. Tickets are available from either the tram driver (enter at the front of the tram or bus) or better from the conductor (enter at the rear part of the tram). Single tickets are fairly expensive. If you plan to make more extensive use of public transport within Amsterdam, there are attractive 24/48/72-hour tickets available from vending machines at Amsterdam Centraal metro station or the GVB office opposite the main entrance of Amsterdam Centraal station. All tickets are interoperable between tram, metro and city buses. See local transport operator GVB for more details. 

The closest tram stop for the conference venue is Koningsplein (King's square) served by tram lines 1, 2 and 5 from Amsterdam Centraal, literally every few minutes. 

Hotel information:

Despite more than 1600 hotels in Amsterdam, finding a good and reasonably priced place to stay is unfortunately not so easy, in particular during tourist season in summer. Early booking and careful reading of room descriptions as well as reports from previous guests is highly recommended to avoid disappointment. The hotel market is extremely volatile and thus it makes little sense to recommend specific places here. Also the range of hotels in easy walking distance from the symposium venue is in the hundreds. We recommend meta booking engines such as trivago for the best offers.

As a rule of thumb, avoid hotels in the red light district. Most of them are pretty shabby and the neighbourhood tends to be very noisy at night. Also be aware that many 3-star hotels in Amsterdam offer basic/economy/budget rooms without private bath room. 

List of accepted papers:

Out of 34 submissions the programme committee selected the following 15 papers for presentation at the symposium:
  1. Usman Dastgeer and Christoph Kessler: 
    Smart containers and skeleton programming for GPU-based systems 
  2. Carlos Alberto Martínez-Angeles, Inês Dutra, Vítor Santos Costa and Jorge Buenabad-Chavez: 
    Relational Learning with Datalog and GPUs: Accelerating Rule Coverage 
  3. Ye Wang and Zhiyuan Li: 
    GridFOR: A Domain Specific Language for Parallel Grid-based Applications 
  4. Alvaro Estebanez, Diego R. Llanos and Arturo Gonzalez-Escribano: 
    New Data Structures to Handle Speculative Parallelization at Runtime
  5. Konrad Siek and Paweł T. Wojciechowski: 
    Atomic RMI: a Distributed Transactional Memory Framework 
  6. Marco Aldinucci, Sonia Campa, Marco Danelutto, Peter Kilpatrick and Massimo Torquati:
    Pool evolution: a domain specific parallel pattern 
  7. Tristan Aubrey-Jones and Bernd Fischer: 
    Synthesizing MPI Implementations from Functional Data-Parallel Programs 
  8. Antoine Tran Tan, Joel Falcou, Daniel Etiemble and Hartmut Kaiser: 
    Automatic Task-based Code Generation for High Performance Domain Specific Embedded Language 
  9. Jean Fortin and Frédéric Gava:
    BSP-Why: a tool for deductive verification of BSP algorithms with subgroup synchronisation 
  10. Miguel Areias and Ricardo Rocha:
    A Lock-Free Hash Trie Design for Concurrent Tabled Logic Programs 
  11. Tarek Menouer, Mohamed Rezgui, Bertrand Le Cun and Jean-Charles Régin:
    Mixing Static and Dynamic Partitionning to Parallelize a Constraint Programming Solver 
  12. Shigeyuki Sato and Kiminori Matsuzaki:
    A Generic Implementation of Tree Skeletons 
  13. Jose M. Andion, Manuel Arenaz, François Bodin, Gabriel Rodríguez and Juan Tourino:
    Locality-Aware Automatic Parallelization for GPGPU with OpenHMPP Directives 
  14. Kiminori Matsuzaki and Reina Miyazaki:
    Parallel Tree Accumulations on MapReduce 
  15. Ali Jannesari, Nico Koprowski, Felix Wolf, Jochen Schimmel and Walter F. Tichy:
    Generating Parallel Unit Tests for Correlated Variables 

    Programme committee:

    • Marco Aldinucci, University of Torino, Italy
    • Jost Berthold, University of Copenhagen, Denmark
    • Rob Bisseling, Utrecht University, Netherlands
    • Murray Cole, University of Edinburgh, United Kingdom
    • Joel Falcou, MetaScale / Université Paris-Sud, France
    • Clemens Grelck, University of Amsterdam, Netherlands (Chair)
    • Gaétan Hains, Université Paris-Est, France
    • Zhenjiang Hu, National Institute of Informatics, Japan
    • Youry Khmelevsky, University of British Columbia (Okanagan) / Okanagan College, Canada
    • Herbert Kuchen, University of Münster, Germany
    • Kiminori Matsuzaki, Kochi University of Technology, Japan
    • Frank Penczek, Intel Ulm, Germany
    • Susanna Pelagatti, University of Pisa, Italy
    • Tiark Rompf, Oracle Labs / Ecole Polytechnique Fédérale de Lausanne, Switzerland
    • Francisco de Sande, University of La Laguna, Spain
    • Kostis Sagonas, Uppsala University, Sweden
    • Vijay Saraswat, IBM Thomas J. Watson Research Center, USA
    • Sven-Bodo Scholz, Heriot-Watt University, United Kingdom

    Organiser and programme chair:

    Clemens Grelck (Universiteit van Amsterdam, Netherlands)

    HLPP steering committee:

    • Clemens Grelck (Universiteit van Amsterdam, Netherlands)
    • Gaétan Hains (Université Paris-Est, France)
    • Kiminori Matsuzaki (Kochi University of Technology, Japan) 
    • Frédéric Loulergue (Université d'Orléans, France) 
    • Quentin Miller (Somerville College Oxford, United Kingdom)
    • Alexander Tiskin (University of Warwick, United Kingdom)

    Previous HLPP symposia and workshops: