DAY 1: January 11th (Wednesday)

08:30 Registration (“Sala della Basilica”)
09:00 Welcome
09:15 Auto-Vectorization Techniques for Modern SIMD Architectures
Olaf Krzikalla, Kim Feldhoff, Ralph Müller-Pfefferkorn, Wolfgang E. Nagel (Technische Universität, Dresden)
Tags: extraction of parallelism, vectorization
09:40 Portable Support for Explicit Vectorisation in C
Artjoms Sinkarovs (University of Hertfordshire), Sven-Bodo Scholz (Heriot-Watt University)
Tags: languages for parallelism, vectorization
10:05 An Evaluation of Vectorizing Compilers *
Saeed Maleki (University of Illinois at Urbana-Champaign), Yaoqing Gao (IBM Toronto Laboratory), María Jesús Garzarán (University of Illinois at Urbana-Champaign), Tommy Wong (IBM Toronto Laboratory), David A. Padua (University of Illinois at Urbana-Champaign)
Tags: loop parallelization, performance evaluation, vectorization
10:30 Coffee break
11:00 A Functional Approach to Optimize SIMD Computations of OpenCL Programs
Yu-Te Lin, Chi-Bang Kuan, Shao-Chung Wang, Jenq-Kuen Lee (National Tsing-Hua University)
Tags: extraction of parallelism, OpenCL, vectorization
11:25 Source-to-Source Vectorization for OpenCL Kernels
Jianbin Fang, Ana Lucia Varbanescu, Alexander S. van Amesfoort, Henk J. Sips (Delft University of Technology)
Tags: extraction of parallelism, GPUs, OpenCL, vectorization
11:50 Automatic Loop Parallelization Using Transactional Memory Support
Miguel Angel Gonzalez-Mesa, Ricardo Quislant, Eladio Gutierrez, Oscar Plata (University of Malaga)
Tags: extraction of parallelism, loop parallelization, parallelizing compilers, transactional memories
12:15 Compiler-Assisted Resource Management for CUDA Programs
Yi-Ping You, Yu-Shiuan Tsai (National Chiao Tung University)
Tags: GPUs, optimizing compilers, register allocation, performance prediction
12:40 Lunch
14:30 Studying Spilling in the Light of SSA
Quentin Colombet, Florian Brandner, Alain Darte (Ecole Normale Supérieure de Lyon)
Tags: optimizing compilers, register allocation, SSA form
14:55 Kernel Offloading with Optimized Remote Accesses
Christophe Alias, Alain Darte, Alexandru Plesco (Ecole Normale Supérieure de Lyon)
Tags: hw accelerators, GPUs, optimizing compilers, polyhedral optimizations
15:20 Performance Modeling for System Selection
Rosario Cammarota (University of California, Irvine), Arun Kejariwal, Paolo D’Alberto (Yahoo!), Alexandru Nicolau (University of California, Irvine), Alexander Veidenbaum (Yahoo!)
Tags: auto-tuning, performance evaluation, performance modeling, performance prediction
15:45 Dynamic Selection of Tile Sizes
Sanket Tavarageri, Louis-Noël Pouchet (Ohio State University), J. Ram Ramanujam (Louisiana State University), Atanas Rountev, P. Saday Sadayappan (Ohio State University)
Tags: auto-tuning, loop optimization, performance evaluation, performance prediction
16:10 Coffee break
16:40 Guided Reverse Analysis for Dynamically Extensible Systems
Jeffery von Ronne, Keyvan Nayyeri (University of Texas at San Antonio)
Tags: performance prediction, parallel runtime systems
17:05 A framework for argument-based task synchronization
Carlos H. González, Basilio B. Fraguela (University of A Coruña)
Tags: extraction of parallelism, parallel runtime systems
17:30 Discussion session: “Benchmarks for Compiler Evaluation” – Part I: questions/ideas (1 hour)
Coordinators: Gianfranco Bilardi (University of Padova), David A. Padua (University of Illinois at Urbana-Champaign)
19:45 Social dinner (Ristorante Belle Parti)


DAY 2: January 12th (Thursday)

09:00 Layer-Unified Coarse Grain Task Parallel Processing for Java Programs
Akimasa Yoshida, Tomohiro Ozawa (Toho University)
Tags: extraction of parallelism, parallelizing compilers
09:25 Fine Granular Parallel Programming Approach for Molecular Dynamics Simulation on Petascale Machines
Haowei Huang, Michael Gerndt (Technische Universität, München)
Tags: applications of parallelism, languages for parallelism
09:50 Using Whole-Program Analysis, Program Slicing, and Loop Transformations to Optimise Unstructured Mesh Applications
Adam Betts, Carlo Bertolli, Paul Kelly (Imperial College London), David Radford (Rolls Royce), Gihan Mudalige, Mike Giles (University of Oxford)
Tags: applications of parallelism, languages for parallelism
10:15 Coffee break
10:45 The Pochoir Stencil Compiler
Yuan Tang (Fudan University and MIT CSAIL), Rezaul Chowdhury (Boston University and MIT CSAIL), Bradley C. Kuszmaul (Tokutek and MIT CSAIL), Chi-Keung Luk (Intel and MIT CSAIL), Charles E. Leiserson (MIT CSAIL)
Tags: extraction of parallelism, languages for parallelism, parallelizing compilers
11:10 OpenCL vs OpenMP: a Programmability Debate
Jie Shen, Jianbin Fang (Delft University of Technology), Ana Lucia Varbanescu (Delft University of Technology and VU University Amsterdam), Henk J. Sips (Delft University of Technology)
Tags: languages for parallelism, OpenCL, OpenMP, performance evaluation
11:35 A New Method for Program Inversion
Cong Hou, George Vulov (Georgia Institute of Technology), Daniel Quinlan, David Jefferson (Lawrence Livermore National Laboratory), Richard Fujimoto, Richard Vuduc (Georgia Institute of Technology)
Tags: debugging, program inversion
12:00 Computing-Kernels Performance Prediction Using DataFlow Analysis and Microbenchmarking
Eric Petit (University of Versailles Saint-Quentin-en-Yvelines), Pablo de Oliveira Castro, Tarek Menouer (Exascale Computing Research), Bettina Krammer, William Jalby (University of Versailles Saint-Quentin-en-Yvelines)
[Submission 08 – Tags: performance metrics, performance prediction]
12:25 Lunch
14:15 Social event: guided tour of Padova monuments (3 hours and 30 minutes)


DAY 3: January 13th (Friday)

09:00 Two Alternative Implementations of Automatic Parallelisation
William Paul Cockshott, Youssef Gdura, Paul Keir (University of Glasgow)
Tags: compilers, extraction of parallelism, vectorization
09:25 A Compiler Extension for Parallelizing Arrays Automatically on the Cell Heterogeneous Processor
Youssef Gdura, William Paul Cockshott (University of Glasgow)
Tags: compilers, extraction of parallelism, vectorization
09:50 Performance-Aware Dynamic Composition of Applications for Heterogeneous Multicore Systems with the PEPPHER Composition Tool
Usman Dastgeer, Lu Li, Christoph W. Kessler (Linköping University)
Tags: extraction of parallelism, GPUs, languages for parallelism, multicores
10:15 Coffee break
10:45 OSCAR Parallelizing Compiler and API for Real-time Low Power Heterogeneous Multicores
Akihiro Hayashi, Mamoru Shimaoka, Hiroki Mikmi, Masayoshi Mase, Yasutaka Wada, Jun Shirako, Keiji Kimura, Hironori Kasahara (Waseda University)
Tags: heterogeneous multicores, parallelizing compilers, power consumption
11:10 The Tuning Language for Concurrent Collections
Kathleen Knobe (Intel), Michael G. Burke (Rice University)
Tags: languages for parallelism
11:35 Synthesizing Concurrent Graph Data Structures: a Case Study
Roman Manevich, Rashid Kaleem, Keshav Pingali (University of Texas at Austin)
Tags: parallel data structures, parallelizing compilers
12:00 Lunch
14:00 Optimizing a D2Q37 Lattice Boltzmann Code: multi-core CPUs vs GP-GPUs
Luca Biferale (University of Rome Tor Vergata and INFN), Filippo Mantovani (University of Regensburg), Marcello Pivanti (University of Ferrara and INFN), Fabio Pozzati (Bruno Kessler Foundation), Mauro Sbragaglia (University of Rome Tor Vergata and INFN), Andrea Scagliarini (University of Barcelona), Sebastiano Fabio Schifano (University of Ferrara and INFN), Federico Toschi (Eindhoven University of Technology), Raffaele Tripiccione (University of Ferrara and INFN)
Tags: applications of parallelism, GPUs, multicores
14:25 Mapping GPU Computing Models to Multicore CPUs: a Case Study
Nasser S. Anssari, John A. Stratton, Wen-mei W. Hwu (University of Illinois at Urbana-Champaign)
Tags: GPUs, multicores, performance evaluation
14:50 OpenMP to OpenCL code generation and optimization *
Dominik Grewe, Zheng Wang, Michael O’Boyle (University of Edinburgh)
Tags: extraction of parallelism, OpenCL, OpenMP
15:15 Coffee break
15:45 Discussion session: “Benchmarks for Compiler Evaluation” – Part II: answers/ideas (1.5 hours)
Coordinators: Gianfranco Bilardi (University of Padova), David A. Padua (University of Illinois at Urbana-Champaign)
17:15 Closing Remarks

* As per request of the authors, the papers marked with an asterisk do not appear in the proceedings of CPC 2012.