|
Compilers for Parallel Computing 2015 | |||
| Time | Title | Authors | Paper# | |
| Wed 7th | ||||
| 10:00 | Opening remarks | |||
| Session chair: | Paul Kelly | |||
| 10:05 | Synthesizing Hot Code Paths by Abductive Reasoning | Simon Moll and Sebastian Hack | 24 | |
| 10:30 | Extending Firedrake to Generate Code for Extruded Meshes | Gheorghe-Teodor Bercea, Andrew T. T. McRae, Fabio Luporini, J. Ramanujan, David A. Ham and Paul H. J. Kelly | 29 | |
| 10:55 | Intelligent Heuristic Construction with Active Learning | William Ogilvie, Pavlos Petoumenos, Zheng Wang and Hugh Leather | 34 | |
| 11:20 | Break | |||
| Session chair: | Sebastian Hack | |||
| 11:45 | Commutativity Characterization of Algorithmic Skeletons and Their Automated Detection in Sequential Legacy Codes | Tobias Edler von Koch and Bjšrn Franke | 5 | |
| 12:10 | Static Analysis for Near-Optimal Thread-Coarsening on GPUs | Nicolai Stawinoga and Anthony Field | 27 | |
| 12:35 | Extending the Roofline Model: Bottleneck Analysis with Microarchitectural Constraints | Victoria Caparr—s Cabezas and Markus Pźschel | 22 | |
| 13:00 | Lunch | |||
| Session chair: | Henk Sips | |||
| 14:00 | Compiler Analysis of Data Movement Lower Bounds for Affine Programs | Venmugil Elango, Fabrice Rastello, Louis-Noel Pouchet, J Ramanujam and P Sadayappan | 48 | |
| 14:25 | From Fortran to C: Extensions for C Source-to-Source Compilers | Nelson Lossing, Pierre Guillou, Mehdi Amini and FranŤois Irigoin | 19 | |
| 14:50 | A Team-based Design Methodology for Memory Hierarchy-Aware Runtime Support in Coarray Fortran | Dounia Khaldi, Shiyao Ge, Deepak Eachempati, Pierre Jouvelot and Barbara Chapman | 30 | |
| 15:15 | Break | |||
| Session chair: | Ram Ramanujam | |||
| 15:50 | Compiler and Runtime Support for Hybrid Static/Dynamic Scheduling | Hans Vandierendonck | 16 | |
| 16:15 | Low Overhead Runtime Code Specializations: A Case Study of the Impact on Speed, Energy and Memory | Victor Lomuller and Henri-Pierre Charles | 18 | |
| 16:40 | Limits of static dependence analysis for automatic parallelization | Niall Murphy, Tim Jones, Simone Campanoni and Robert Mullins | 32 | |
| 17:05 | ||||
| 18:00 | Refreshments | Provided by the Data Science Institute | Huxley Building, 180 Queen's Gate | |
| 18:30 | Public lecture | Graph Analytics on the Galois System | Keshav Pingali (Clore theatre, Huxley Building) | |
| http://www3.imperial.ac.uk/data-science/events#pingali | ||||
| Dinner on your own | ||||
| Thu 8th | ||||
| Session chair: | Francois Irigoin | |||
| 09:30 | Collective Mind, Part II: Towards Performance- and Cost-Aware Software Engineering as a Natural Science | Grigori Fursin, Abdul Memon, Christophe Guillon and Anton Lokhmotov | 2 | |
| 09:55 | A Mixture of Experts Approach for Parallelism Mapping in Dynamic Environments | Murali Emani and Michael O'Boyle | 31 | |
| 10:20 | A Basic Linear Algebra Compiler | Daniele Spampinato and Markus Pźschel | 13 | |
| 10:45 | Break | |||
| Session chair: | David Ham | |||
| 11:20 | Generating Optimized Fourier Interpolation Routines for Density Functional Theory using SPIRAL | Doru Thom Popovici, Francis P Russell, Karl Wilkinson, Chris-Kriton Skylaris, Paul H. J. Kelly and Franz Franchetti | 42 | |
| 11:45 | The ICON Domain Specific Language | Leonidas Linardakis, Julian Kunkel and Ulrich Koerner | 14 | |
| 12:10 | Generating OpenCL Code from PENCIL | Sven van Haastregt, Alexey Kravets, Vincent Grevendonk andAnton Lokhmotov | 36 | |
| 12:35 | Lunch | |||
| Session chair: | Thomas Gross | |||
| 13:35 | Keynote talk | Elixir: A System for Synthesizing Parallel Graph Analytics Programs | Keshav Pingali | |
| 14:20 | Nessie: A NESL to CUDA Compiler | John Reppy and Nora Sandler | 15 | |
| 14:45 | Program Models of Footprint in Shared Cache | Chen Ding, Hao Luo and Chencheng Ye | 6 | |
| 15:10 | Break | |||
| Free afternoon | We invite you to explore the South Kensington museums, which are all free admission, and within a very few minutes walk | |||
| Dinner at 7 | ||||
| Fri 9th | ||||
| Session chair: | Hans Vandierendonck | |||
| 09:30 | Irrevocable Transactions to Handle Signature Saturation | Ricardo Quislant, Eladio GutiŽrrez and Oscar Plata | 25 | |
| 09:55 | Compiling Vector Pascal to the XeonPhi | Paul Cockshott, Mozhgan Chimeh, Susanne Oehler, Ashkan Tousimojarad and Tian Xu | 3 | |
| 10:20 | Code generation for distributed- and shared-memory hierarchical systems with Trasgo | Ana Moret—n, Arturo Gonzalez-Escribano and Diego R. Llanos | 21 | |
| 10:45 | Break | |||
| Session chair: | John Reppy | |||
| 11:20 | Detecting Anomalies in Concurrent Programs based on Dynamic Control Flow Changes | Faheem Ullah and Thomas R. Gross | 9 | |
| 11:45 | A Prediction-based Method for Deploying Applications on Heterogeneous Platforms | Jie Shen, Ana Lucia Varbanescu and Henk Sips | 28 | |
| 12:10 | A Financial Benchmark for GPGPU Compilation | Christian Andreetta, Jost Berthold, Martin Elsman and Cosmin Oancea | 37 | |
| 12:35 | Lunch | |||
| Session chair: | Luigi Nardi | |||
| 13:45 | Bulk-Synchronous Communication Mechanisms in Diderot | John Reppy and Lamont Samuels | 23 | |
| 14:10 | The Paralldroid Project, Models and Cases of Use | Alejandro Acosta and Francisco Almeida | 35 | |
| 14:35 | Chaining the RIPL and SaC DSLs for Image Processing | Robert Stewart, Deepayan Bhowmik, Greg Michaelson and Sven-Bodo Scholz | 41 | |
| 15:00 | Break | |||
| Session chair: | Paul Kelly | |||
| 15:15 | Towards a Multitarget C and OpenCL Generation from MATLAB | Lu’s Reis, Jo‹o Bispo and Joao Cardoso | 45 | |
| 15:40 | Nesoi: Compile time checking of transactional coverage in parallel programs | Daniel Goodman, Behram Khan, Mikel Lujan and Ian Watson | 4 | |
| 16:05 | Close | |||