LCPC2006: Preliminary Program

Thursday Friday Saturday

Thursday, Nov 2nd

8:30-10:00 - Session 1: Keynote: Compilation Techniques for Partitioned Global Address Space Languages [PDF Slides]
Speaker: Kathy Yelick, UC Berkeley and Lawrence Berkeley National Lab

10:00-10:30 - Coffee Break

10:30-12:00 - Session 2: Programming Models

Can Transactions Enhance Parallel Programs? [Full paper] [Slides]
Troy A. Johnson, Sang-Ik Lee, Seung-Jai Min and Rudolf Eigenmann

Design and Use of htalib -- a Library for Hierarchically Tiled Arrays [Full paper] [Slides]
Ganesh Bikshandi, Jia Guo, Christoph von Praun, Gabriel Tanase, Basilio B. Fraguela, Maria J. Garzaran, David Padua and Lawrence Rauchwerger

SP@CE - An SP-based Programming Model for Consumer Electronics Streaming Applications [Full paper] [Slides]
Ana Lucia Varbanescu, Maik Nijhuis, Arturo Gonzalez Escribano, Henk Sips, Herbert Bos and Henri Bal

12:00-13:00 - Lunch

13:00-15:00 - Session 3: Code generation
Chair: Peng Wu, IBM Research

Data Pipeline Optimization for Shared Memory Multiple-SIMD Architecture [Full paper]
Weihua Zhang, Tao Bao, Binyu Zang and Chuanqi Zhu

Dependence-based Code Generation for a CELL Processor [Full paper] [Slides]
Yuan Zhao and Ken Kennedy

Expression and Loop Libraries for High-Performance Code Synthesis [Full paper]
Christopher Mueller and Andrew Lumsdaine

Specializing Code for FFT libraries [Full paper] [Slides]
Minhaj KHAN and Henri-Pierre CHARLES

15:00-15:30 - Coffee Break

15:30-17:00 - Session 4: Parallelism

A Characterization of Shared Data Access Patterns in UPC Programs [Full paper] [Slides]
Christopher Barton, Calin Cascaval and Jose Amaral

Exploiting Speculative Thread-Level Parallelism in Data Compression Applications [Full paper] [Slides]
Shengyue Wang, Antonia Zhai and Pen-Chung Yew

On Control Signals for Multidimensional Time [Full paper] [Slides]
DaeGon Kim, Gautam Gupta and Sanjay Rajopadhye

Friday, Nov 3rd

8:30-10:00 - Session 5: Keynote: The Berkeley View: A New Framework and a New Platform for Parallel Research [Slides]
Speaker: David Patterson, UC Berkeley

10:00-10:30 - Coffee Break

10:30-12:00 - Session 6: Compilation Techniques

An Effective Heuristic for Simple Offset Assignment with Variable Coalescing [Full paper] [Slides]
Hassan Salamy and J. (Ram) Ramanujam

Iterative Compilation by Exploration of Kernel Decomposition [Full paper] [Slides]
Denis Barthou, Sebastien Donadio, Alexandre Duchateau, William Jalby and Eric Courtois

Quantifying Uncertainty in Points-To Relations [Full paper] [Slides]
Constantino Ribeiro and Marcelo Cintra

12:00-13:00 - Lunch

13:00-14:30 - Session 7: Data Structures

Cache Behavior Modelling for Codes Involving Banded Matrices [Full paper] [Slides]
Diego Andrade, Basilio B. Fraguela and Ramón Doallo

Tree-Traversal Orientation Analysis [Full paper] [Slides]
Kevin Andrusky, Stephen Curial and Jose Nelson Amaral

UTS: An Unbalanced Tree Search Benchmark [Full paper] [Slides]
Stephen Olivier, Jun Huan, Jinze Liu, Jan Prins, James Dinan, P. Sadayappan and Chau-Wen Tseng

14:30-15:00 - Coffee Break

15:00-17:00 - Session 8: Register Allocation

Copy Propagation Optimizations for VLIW DSP Processors with Distributed Register Files [Full paper] [Slides]
Jenq-Kuen Lee, Sheng-Yuan Chen and Chung-Ju Wu

Optimal Bitwise Register Allocation using Integer Linear Programming [Full paper] [Slides]
Rajkishore Barik, Christian Grothoff, Rahul Gupta, Vinayaka Pandit and Raghavendra Udupa

Register Allocation: What does the NP-Completeness Proof of Chaitin et al. Really Prove? [Full paper] [Slides]
Florent Bouchez, Alain Darte, Christophe Guillon and Fabrice Rastello

19:00-21:00 - Banquet

Guest speaker: Dr. Frederica Darema, Senior Science and Technology Advisor, National Science Foundation [Slides]

Saturday, Nov 4th

9:00-11:00 - Session 9: Memory Management
Chair: Rudolf Eigenmann, Purdue University

Custom Memory Allocation for Free [Full paper] [Slides]
Alin Jula and Lawrence Rauchwerger

Optimizing the use of static buffers for DMA on a CELL chip [Full paper]
Tong Chen, Zehra Sura, Kathryn O'Brien and Kevin O'Brien

Runtime Address Space Computation for SDSM Systems [Full paper] [Slides]
Jairo Balart, Marc Gonzalez, Xavier Martorell, Eduard Ayguade and Jesus Labarta

Static Heap Analysis for Automatic Parallelization [Full paper] [Slides]
Mark Marron, Deepak Kapur, Darko Stefanovic and Manuel Hermenegildo

11:00-12:00 - Closing Remarks