Symposium on Application Accelerators in High-Performance Computing (SAAHPC)

Practical FPGA acceleration for Software Programmers: A tutorial on writing Mitrion-C for the Mitrion Virtual Processor

Open for free to all (would be) users of high-performance computing on FPGAs.

Pre-requisites:

• Basic knowledge of parallel systems
• Basic familiarity with C-family language syntax (C/C++/Java/C#/etc)
• Bring a laptop and mouse with mouse-wheel

For more information, see the tutorial page.

To register for the tutorial, see the tutorial registration page.

For other information regarding the tutorial day, please contact: saahpc_register@mitrionics.com.

Opening Session
Chair: Wen-mei Hwu, UIUC

Danny Powell, NCSA Executive Director

Ravi K. Iyer, Interim Vice Chancellor for Research, UIUC

Volodymyr Kindratenko, SAAHPC technical program chair

Keynote: Massive Data Computing

Heterogeneous Multiprocessing
Chair: Volodymyr Kindratenko, NCSA

Bringing heterogeneous multiprocessors into the mainstream
Model for Programming Heterogeneous Systems (paper also available)
Implementing 3D SPHARM Surfaces Registration on Cell Processor (paper also available)

GPU Programming
Chair: David Kaeli, Northeastern University

Compiler Support for High-level GPU Programming (paper also available)
Decoupled Access/Execute Metaprogramming for GPU-Accelerated Systems (paper also available)
Programmability: Design Costs and Payoffs using AMD GPU Streaming Languages and Traditional Multi-Core Libraries (paper also available)

Application Accelerator Technologies
Chair: Craig Steffen, NCSA

Investigations of Multicore and Acceleration Technologies at ARSC
Three Steps to Accelerator Success
The golden triangle of Hybrid Computing: FPGAs, GPGPUs and Many-Core processors

Reception and Poster Session

The PetaFlops Router: Harnessing FPGAs and Accelerators for High Performance Computing - Z. Baker, T. Bhattacharya, M. Dunham, P. Graham, R. Gupta, J. Inman, A. Klein, G. Kunde, A. McPherson, M. Stettler, and J. Tripp

A New Coarse-Grained Reconfigurable Architecture with Fast Data Relay and Its Compilation Flow - Lu Wan, Chen Dong, and Deming Chen (paper also available)

Accelerating Particle Image Velocimetry Using Hybrid Architectures - Vivek Venugopal, Cameron Patterson, and Kevin Shinpaugh (paper also available)

Can We Apply Accelerator-Cores to Control-Intensive Programs? - Sean Rul, Hans Vandierendonck, and Koen De Bosschere

Data parallel loop statement extension to CUDA: GpuC - Zeki Bozkus, Rajeev Thakur, William Gropp, and Ewing Lusk

GPU Acceleration of Equations Assembly in Finite Elements Method - Preliminary Results - Jiri Filipovic Igor Peterlik, and Jan Fousekz (paper also available)

In-Socket FPGA Implementation of Bioinformatic Algorithms Using the Intel AAL - Vincent Natoli, Jeff Allred, Jack Coyne, and William Lynch

A massively parallel framework using P systems and GPUs - Jose Cecilia, Gines Guerrero, Jose Garcia, Miguel Martinez-del-Amor, Ignacio Perez-Hurtado, and Mario Perez-Jimenez (paper also available)

Breaking the Sequential Dependency and Extracting Parallelism out of Recurrence Equations—Sample Application to Options Pricing in Finance and HMMER in Computational Biology - Narayan Ganesan, Roger Chamberlain, and Jeremy Buhler (paper also available)

Accelerating a Software Radio Astronomy Correlator using FPGA co-processors - Andrew Woods, Michael Inggs, and Alan Langman (paper also available)

Accelerating the Stochastic Simulation Algorithm - David Jenkins and Gregory Peterson (paper also available)

Assembling Large Mosaics of Electron Microscope Images using GPU - Kannan Venkataraju, Mark Kim, Dan Gerszewski, James R. Anderson, and Mary Hall (paper also available)

Abstraction Library To Simplify Application FPGA Configuration and Control - Craig Steffen

An Automated Approach for SIMD Kernel Generation for GPU based Software Acceleration - Kanupriya Gulati and Sunil Khatri

Debugging CUDA with TotalView - Chris Gottbrath

Accelerated Root Finding for Computational Finance - Mark Bennett (paper also available)

Evaluating the Jaccard-Tanimoto Index on Multi-Core Architectures - Vipin Sachdeva, Douglas Freimuth, and Chris Mueller

Keynote II
Chair: Rob Pennington, NSF

Keynote: Accelerating past the petascale
Software Acceleration — Optimizing for multi-core

GPU Computing
Chair: Greg Newby, Arctic Region Supercomputing Center

Raising the Level of GPU Computing
On Testing GPU Memory for Hard and Soft Errors (paper also available)
Software-Based ECC for GPUs (paper also available)

Reconfigurable Computing Applications
Chair: Eric Stahlberg, Wittenberg University/OpenFPGA

Reconfigurability, Reliability and Standards: A View from the Frontlines
Production Floating Point Applications on FPGAs (paper also available)
Mixed Precision Dense Linear System Solvers for High Performance Reconfigurable Computing (paper also available)
Reconfigurable Computing for Cholesky Decomposition (paper also available)

GPU Applications
Chair: John Stone, University of Illinois

Accelerating the ANSYS Direct Sparse Solver with GPUs (paper also available)
Accelerating Free-Energy Minimization using Graphics Processors (paper also available)
GPU Acceleration of the Generalized Interpolation Material Point Method (paper also available)

Optional tour of NCSA computing facilities

OpenFPGA Session
Chair: Eric Stahlberg, Wittenberg University/OpenFPGA

GPU Applications
Chair: Gregory Peterson, University of Tennessee

Flexible Hardware Mapping for Finite Element Simulations on Hybrid CPU/GPU Clusters (paper also available)
Architectural Comparisons for a Quantum Monte Carlo Application (paper also available)
GPU Coprocessing for Wireless Network Simulation (paper also available)

Novel Architectures
Chair: Volodymyr Kindratenko, NCSA

Rigel: A Scalable Architecture for 1,000+ Core Accelerators (paper also available)
Reconfigurable Active Drive: An FPGA Accelerated Storage Architecture for Data-Intensive Applications (paper also available)
Extending a Stream Programming Paradigm to Hardware Accelerator Platforms (paper also available)

Optional tour of NCSA computing facilities