MTAAP 2010

Workshop on Multithreaded Architectures and Applications

http://ft.ornl.gov/events/mtaap10/

To be held in Conjunction with the International Parallel and Distributed Processing Symposium (IPDPS 2010) April 23, 2010, Atlanta

4/20/2010 - New agenda posted to accommodate Eyjafjallajokull travel disruption.

Theme

Multithreading (MT) programming and execution models are starting to permeate the high-end and mainstream computing scene. This trend is driven by the need to increase processor utilization and deal with the memory-processor speed gap. Recent and upcoming examples architectures that fit this profile are Cray's XMT, IBM Cyclops, and several SMT processors from Sun (Victoria Falls), IBM (Power6, Power7), or Intel, as well as heterogeneous systems with accelerators such as GPGPUs from ATI, NVIDIA, and Intel. The underlying rationale to increase processor utilization is a varying mix of new metrics that take performance improvements as well as better power and cost budgeting into account. Yet, it remains a challenge to identify and productively program applications for these architectures with a resulting substantial performance improvement.

The MTAAP 2010 workshop is a full-day meeting to be held at the IPDPS 2010 focusing on Multithreading architectures and applications. This workshop intends to identify applications that are amenable to MT and the MT programming and execution models as well as the underlying architectures on which they can thrive. The workshop seeks to explore programming frameworks in the form of languages and libraries, compilers, analysis and debugging tools to increase the programming productivity. The topics of interest, of both theoretical and practical significance, include but are not limited to:

- Multithreaded Architectures

- Multithreaded Programming Framework

- Compilation and Optimization for MT architectures

- Multithreaded Performance Analysis and Debugging Tools

- Multithreaded Performance Metrics and Evaluations

- Multithreaded Libraries and run-time systems

- Innovative applications for MT architectures

The MTAAP workshop proceedings will be published along with the IPDPS proceedings.

Workshop Presentations

Workshop presentations are available at http://ft.ornl.gov/events/mtaap10/presentations/.

Workshop Agenda for Friday, April 23

4/20/2010 - SCHEDULE ALERT: The difficulties of traveling from Europe caused by the Eyjafjallajokull eruption have forced us to rearrange the agenda. Please review the new agenda below, and alert the workshop chairs to any travel problems or issues with your new time slot.

08:00 - 08:15 : MTAAP 2010 Welcome

08:15 - 10:15 : Systems, Tools, and Applications I

• On the Parallelisation of of MCMC by Speculative Chain Execution: Jonathan M R Byrd (University of Warwick, United Kingdom); Stephen Jarvis (University of Warwick, United Kingdom)
• User Level DB: a Debugging API for User-Level Thread Libraries: Kevin Pouget (CEA/DAM Ile de France, France); Marc Perache (CEA/DAM Ile de France, France); Patrick Carribault (CEA/DAM Ile de France, France); Herve Jourdren (CEA/DAM Ile de France, France)
• A Multi-Threaded Approach for Data-Flow Analysis: Marcus Edvinsson (Vaxjo University, Sweden); Welf Lowe (Vaxjo University, Sweden)
• Scheduling complex streaming applications on the Cell processor: Matthieu Gallet (Ecole normale superieure de Lyon, France); Mathias Jacquelin (ENS Lyon, France); Loris Marchal (CNRS, France)

10:15 - 10:45 : Break I

10:45 - 11:45 : Keynote

• The POWER7 Processor
• Dr. Ram Rajamony, IBM Austin Research Laboratory
• Abstract: IBM recently announced the POWER7 systems that featuring an innovative multi-core, 45 nanometer design, running at speeds of over 4.1 GHz, with up to 8-cores per socket, and four threads per core. These processors are organized into a range of systems that handle workloads ranging from transactional to high-performance computing workloads. This talk will describe the architecture of the POWER7 processor and systems, with special emphasis on the multi-core and multi-threading aspects.

11:45 - 13:00 : Lunch

13:00 - 15:00 : Systems, Tools, and Applications II

• Modeling Bounds on Migration Overhead for a Traveling Thread Architecture: Patrick La Fratta (University of Notre Dame, USA); Peter Kogge (University of Notre Dame, USA)
• TiNy Threads on BlueGene/P: Exploring Many-Core Parallelisms Beyond The Traditional OS: Handong Ye (University of Delaware, USA); Robert S Pavel (University of Delaware, USA); Aaron Landwehr (University of Delaware, USA); Guang Gao (Univ. of Delaware, USA)
• Experimental Comparison of Emulated Lock-free vs. Fine-grain Locked Data Structures on the Cray XMT: Rob Farber (Pacific Northwest Laboratory, USA); David W. Mizell (Cray, USA)
• Large Scale Complex Network Analysis using the Hybrid Combination of a MapReduce cluster and a Highly Multithreaded System: Seunghwa Kang (Georgia Institute of Technology, USA); David A. Bader (Georgia Institute of Technology, USA)

15:00 - 15:30 : Break II

15:30 - 17:00 : Programming Framework

• Out-of-Core Distribution Sort in the FG Programming Environment: Priya Natarajan (Dartmouth College, USA); Thomas H. Cormen (Dartmouth College, USA); Elena Riccio Strange (A9.com, USA)
• Massive Streaming Data Analytics: A Case Study with Clustering Coefficients: David Ediger (Georgia Institute of Technology, USA); Karl Jiang (Georgia Institute of Technology, USA); Jason Riedy (Georgia Institute of Technology, USA); David A. Bader (Georgia Institute of Technology, USA)
• Hashing Strategies for the Cray XMT: Eric Goodman (Sandia National Laboratories, USA); David J Haglin (Pacific Northwest National Laboratory, USA); Chad Scherrer (Pacific Northwest National Laboratory, USA); Daniel Gerardo Chavarria (Pacific Northwest National Laboratory, USA); Jace Mogill (Pacific Northwest National Laboratory, USA); John Feo (Pacific Northwest National Laboratory, USA)

17:00 : Adjourn

Workshop Organization

Chairs

- Luiz DeRose (Cray) (ldr@cray.com)

- Jeffrey Vetter (ORNL and Georgia Tech) (vetter@computer.org)

Program Committee

- David Bader (Georgia Tech)

- Jonathan Berry (Sandia National Laboratory)

- Daniel Chavarria (Pacific Northwest National Laboratory)

- John Feo (Pacific Northwest National Laboratory)

- Guang Gao (U. Delaware)

- Bruce Hendrickson (Sandia National Laboratory)

- Larry Kaplan (Cray)

- Peter Kogge (Notre Dame)

- Michael Merrill (DoD)

- Jose Moreira (IBM)

- P. Sadyappan (Ohio State)

- Mateo Valero (Universitat Politecnica de Catalunya)

- Hans Zima (NASA Jet Propulsion Laboratory)

Proceedings

The proceedings of this workshop will be published together with the proceedings of other IPDPS 2010 workshops by the IEEE Computer Society Press.

Workshop Archive

Information and papers from the 2009 MTAAP workshop is available at http://www.cs.umd.edu/~als/MTAAP09/MTAAP09.html.

Call for Papers

Paper Submission Guidelines

UPDATED TO CONFORM W/ IPDPS GUIDELINES: Submitted manuscripts may not exceed 15 single-spaced pages using 12-point size font on 8.5x11 inch pages, including figures, tables, and references. Please use the standard 1-inch margin. Authors may submit additional material as an appendix to their submission, but there is no guarantee that this material will influence the review process.

Submitted manuscripts may not exceed 20 single-spaced pages using a 12-point font on standard 8.5 by 11-inch pages, everything included (figures, tables, references, etc.).

Manuscripts must be submitted electronically and in either PostScript or PDF format. Submissions will be judged on correctness, originality, technical strength, significance, quality of presentation, and interest and relevance to the workshop attendees. Submitted papers may not have appeared in or be under consideration for another workshop, conference, or journal.

MTAAP submissions are being handled by EDAS system. To submit a paper, please use the following link (MTAAP EDAS) and follow the instructions.

Important Dates

- Papers due: 13 Dec 2009; extended to 20 Dec 2009

- Notification of acceptance: 11 Jan 2010

- Camera-ready due: 1 Feb 2010

Additional Information

E-mail Contact

For more information on MTAAP 2010 or if you have any questions please contact the workshop chairs (e-mail addresses listed above).

 This website is hosted by the ORNL Future Technologies Group.