High-Performance Computing Using FPGAs

This book provides an overview of architectures, tools and applications for High-Performance Reconfigurable Computing . It presents cases from several application domains where HPRC has been used successfully.

High-Performance Computing using FPGA covers the area of high performance reconfigurable computing (HPRC). This book provides an overview of architectures, tools and applications for High-Performance Reconfigurable Computing (HPRC). FPGAs offer very high I/O bandwidth and fine-grained, custom and flexible parallelism and with the ever-increasing computational needs coupled with the frequency/power wall, the increasing maturity and capabilities of FPGAs, and the advent of multicore processors which has caused the acceptance of parallel computational models. The Part on architectures will introduce different FPGA-based HPC platforms: attached co-processor HPRC architectures such as the CHREC's Novo-G and EPCC's Maxwell systems; tightly coupled HRPC architectures, e.g. the Convey hybrid-core computer; reconfigurably networked HPRC architectures, e.g. the QPACE system, and standalone HPRC architectures such as EPFL's CONFETTI system. The Part on Tools will focus on high-level programming approaches for HPRC, with chapters on C-to-Gate tools (such as Impulse-C, AutoESL, Handel-C, MORA-C++); Graphical tools (MATLAB-Simulink, NI LabVIEW); Domain-specific languages, languages for heterogeneous computing(for example OpenCL, Microsoft's Kiwi and Alchemy projects). The part on Applications will present case from several application domains where HPRC has been used successfully, such as Bioinformatics and Computational Biology; Financial Computing; Stencil computations; Information retrieval; Lattice QCD; Astrophysics simulations; Weather and climate modeling.

Presents a taxonomy of existing high performance reconfigurable computer architectures Examines the software tools used in the design and programming of HPRC systems Discusses the future of HPRC and speculates on the shape of future HPRC systems Includes supplementary material: sn.pub/extras

Auteur
Wim Vanderbauwhede is currently a Lecturer at the Department of Computing Science of the University of Glasgow.
Dr. Benkrid is currently a Senior Lecturer at School of Engineering and Electronics at The University of Edinburgh.

Texte du rabat

This book is concerned with the emerging field of High Performance Reconfigurable Computing (HPRC), which aims to harness the high performance and relative low power of reconfigurable hardwarein the form Field Programmable Gate Arrays (FPGAs)in High Performance Computing (HPC) applications. It presents the latest developments in this field from applications, architecture, and tools and methodologies points of view. We hope that this work will form a reference for existing researchers in the field, and entice new researchers and developers to join the HPRC community.

The book includes:

• Thirteen application chapters which present the most important application areas tackled by high performance reconfigurable computers, namely: financial computing, bioinformatics and computational biology, data search and processing, stencil computation e.g. computational fluid dynamics and seismic modeling, cryptanalysis, astronomical N-body simulation, and circuit simulation.
• Seven architecture chapters which present both commercial and academic parallel FPGA architectures, low latency and high performance FPGA-based networks and memory architectures for parallel machines, and a high speed optical dynamic reconfiguration mechanism for HPRC.
• Five tools and methodologies chapters which address the important issue of productivity and high performance in HPRC. These include a study of precision and arithmetic issues in HPRC, comparative studies of C-based high level synthesis tools and RTL-based approaches, taxonomy of HPRC tools and a framework of their analysis, and an integrated hardware-software-application approach to HPRC.

Contenu
High-Performance Hardware Acceleration of Asset Simulations.- Monte Carlo Simulation based Financial Computing on the Maxwell FPGA Parallel Machine.- Bioinformatics Applications on the FPGA-based High-Performance Computer RIVYERA.- FPGA-Accelerated Molecular Dynamics.- FPGA-based HPRC for Bioinformatics Applications.- High-Performance Computing for Neuroinformatics using FPGA.- High-Performance FPGA-Accelerated Real-time Search.- High-Performance Data Processing over N-ary Trees.- FPGA-based Systolic Computational-Memory Array for Scalable Stencil Computations.- High performance implementation of RTM seismic modeling on FPGAs: architecture, arithmetic and power issues.- High-Performance Cryptanalysis on RIVYERA and COPACOBANA Computing Systems.- FPGA-based HPRC Systems for Scientific Applications.- Accelerating the SPICE Circuit Simulator using an FPGA - A Case Study.- The Convey Hybrid-Core Architecture.- Low Cost High Performance Reconfigurable Computing.- An FPGA-based supercomputer for statistical physics: the weird case of Janus.- Accelerate Communication, not Computation!.- High-speed torus interconnect using FPGAs.- MEMSCALE: Re-architecting memory resources for clusters.- High-performance computing based on high-speed dynamic reconfiguration.- Reconfigurable arithmetic for HPC.- Acceleration of the Discrete Element Method: From RTL to C-Based Design.- Optimising Euroben Kernels on Maxwell.- Assessing Productivity of High-Level Design Methodologies for High-Performance Reconfigurable Computers.- Maximum performance computing with dataflow engines.