DF21500 Multicore Computing

Papers for Student Presentations (23/3/2011)

Presentation day: 23/3, 09:00 (sharp!) - 17:00, Alan Turing.
Presentations take 20 minutes, plus 5 minutes for opposition/discussion.

The following 13 papers have been selected for presentation:

Session 1: PARALLEL ALGORITHMS I

• 09:00-09:20
  A. Azevedo et al.:
  Parallel H.264 Decoding on an Embedded Multicore Processor.
  Proc. HiPEAC'09 conference, Jan. 2009

  Presenter: Murat Posluk; Opponent: Lan Dong

• 09:25-09:45
  D. Cederman, Ph. Tsigas:
  GPU-Quicksort: A practical Quicksort algorithm for graphics processors.
  ACM Journal of Experimental Algorithmics, Vol. 14, Article No. 1.4, July 2009.

  Presenter: Lan Dong; Opponent: Murat Posluk

• 09:50-10:10
  N. Leischner, V. Osipov, P. Sanders:
  GPU Sample Sort.
  Proc. IPDPS-2010, April 2010.

  Presenter: Breeta SenGupta; Opponent: Dimitar Nikolov

10:10-10:30 Coffee break

Session 2a: PARALLEL ALGORITHMS II

• 10:30-10:50 Scarpazza, Villa, Petrini:
  Efficient Breadth-First Search on the Cell/BE Processor
  IEEE Trans. Par. and Distr. Syst. 19(10):1381-1395, 2008.

  Presenter: Dimitar Nikolov; Opponent: Rikard Holsmark

• 10:55-11:15
  Gary J. Katz and Joseph T. Kider, Jr:
  All-pairs shortest-paths for large graphs on the GPU
  23rd ACM SIGGRAPH/EUROGRAPHICS symposium on Graphics hardware, 2008.

  Presenter: Lars Olofsson; Opponent: Laurent Deloisieres

Session 2b: PERFORMANCE ANALYSIS

• 11:20-11:40
  S. Williams, A. Waterman, D. Patterson:
  Roofline: an insightful visual performance model for multicore architectures.
  Comm. ACM 52(4), April 2009.

  Presenter: Jan Kasper Martinsen; Opponent: ---

Session 3a: GPU PROGRAMMING:

• 13:15-13:35
  T. D. Han and T. Abdelrahman:
  hiCUDA: High-Level GPGPU Programming.
  IEEE Trans. Par. Distr. Syst. 22(1), Jan. 2011.

  Presenter: Danish Jabbar; Opponent: Lars Olofsson

• 13:40-14:00
  Q. Hou et al.:
  BSGP: Bulk-synchronous GPU programming.
  ACM Trans. Graph. 27(3), article 19, 2008.

  Presenter: Laurent Deloisieres; Opponent: Usman Dastgeer.

Session 3b: SCHEDULING:

• 14:05-14:25
  Cédric Augonnet, Samuel Thibault, and Raymond Namyst.
  Automatic Calibration of Performance Models on Heterogeneous Multicore Architectures.
  In Proceedings of the International Euro-Par Workshops 2009, HPPC'09, volume 6043 of Lecture Notes in Computer Science, Delft, The Netherlands, pages 56-65, August 2009. Springer.

  Presenter: Usman Dastgeer; Opponent: Danish Jabbar.

• 14:30-14:50
  H. Park et al.:
  Edge-centric modulo scheduling for coarse-grained reconfigurable architectures.
  Proc. 17th int. conference on Parallel architectures and compilation techniques (PACT), 2008.

  Presenter: Rikard Holsmark; Opponent: Robert Braun

15:00-15:30 Coffee break

Session 4a: AUTOTUNING FOR MULTICORE

• 15:30-15:50
  B. Jang et al.
  Exploiting Memory Access Patterns to Improve Memory Performance in Data-Parallel Architectures.
  IEEE Trans. Par. Distr. Syst. 22(1), Jan. 2011.

  Presenter: Robert Braun; Opponent: Nicolas Melot

Session 4b: Other Optimizations and Analyses

• 15:55-16:15
  G. Chen and P. Stenström:
  A Methodology for Diagnosing Critical Section Bottlenecks in Multithreaded Applications.
  Proc. MULTIPROG-2011 workshop, Heraklion, Jan. 2011, pp. 21-35.

  Presenter: Nicolas Melot; Opponent: Breeta SenGupta

Task: Prepare a 20 minutes presentation of your chosen paper and at least 3 questions on the other paper for opposition.
After the presentation, hand in a written summary of your presented paper on 2-3 pages.

Please send me your presentation slides (ppt or pdf) for approval at least 48h before your presentation. If you do not get any reply, you can proceed and present.