Configuration management in multi-context reconfigurable systems for simultaneous performance and power optimizations

Authors:
Rafael Maestre;Milagros Fernandez;Fadi J. Kurdahi;Nader Bagherzadeh;Hartej Singh
Affiliations:
Departamento de Arquitectura de Computadores, y Automática, Universidad Complutense de Madrid, SPAIN, http://www.dacya.ucm.es/maestre;Departamento de Arquitectura de Computadores, y Automática, Universidad Complutense de Madrid, SPAIN, http://www.dacya.ucm.es/maestre;Department of Electrical and Computer Engineering, University of California, Irvine, USA;Department of Electrical and Computer Engineering, University of California, Irvine, USA;Department of Electrical and Computer Engineering, University of California, Irvine, USA
Venue:
ISSS '00 Proceedings of the 13th international symposium on System synthesis
Year:
2000

Citing 8
Cited 8

Configuration prefetch for single context reconfigurable coprocessors

FPGA '98 Proceedings of the 1998 ACM/SIGDA sixth international symposium on Field programmable gate arrays
Kernel scheduling in reconfigurable computing

DATE '99 Proceedings of the conference on Design, automation and test in Europe
Temporal partitioning combined with design space exploration for latency minimization of run-time reconfigured designs

DATE '99 Proceedings of the conference on Design, automation and test in Europe
MorphoSys: case study of a reconfigurable computing system targeting multimedia applications

Proceedings of the 37th Annual Design Automation Conference
The MorphoSys Parallel Reconfigurable System

Euro-Par '99 Proceedings of the 5th International Euro-Par Conference on Parallel Processing
Architectural Synthesis Techniques for Dynamically Reconfigurable Logic

FPL '96 Proceedings of the 6th International Workshop on Field-Programmable Logic, Smart Applications, New Paradigms and Compilers
Temporal Partitioning and Scheduling for Reconfigurable Computing

FCCM '98 Proceedings of the IEEE Symposium on FPGAs for Custom Computing Machines
A Framework for Scheduling and Context Allocation in Reconfigurable Computing

Proceedings of the 12th international symposium on System synthesis

System-level power-performance trade-offs in task scheduling for dynamically reconfigurable architectures

Proceedings of the 2003 international conference on Compilers, architecture and synthesis for embedded systems
Specific scheduling support to minimize the reconfiguration overhead of dynamically reconfigurable hardware

Proceedings of the 41st annual Design Automation Conference
Low Energy Data Management for Different On-Chip Memory Levels in Multi-Context Reconfigurable Architectures

DATE '03 Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
A Hybrid Prefetch Scheduling Heuristic to Minimize at Run-Time the Reconfiguration Overhead of Dynamically Reconfigurable Hardware

Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
A Reconfiguration Manager for Dynamically Reconfigurable Hardware

IEEE Design & Test
A configuration memory hierarchy for fast reconfiguration with reduced energy consumption overhead

IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
System-level power-performance tradeoffs for reconfigurable computing

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Design and analysis of adaptive processor

ACM Transactions on Reconfigurable Technology and Systems (TRETS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we present a novel solution to the problem of configuration management for multi-context reconfigurable systems targeting DSP applications, its goal being to minimize both, configuration latency and power consumption. We assume that this technique is applied within a larger compilation framework, which provides a scheduled task sequence of the considered application. Reconfiguration latency reduction is the first criteria to consider, and we prove that the optimal solution can be obtained in all cases. Secondly, power is optimized without affecting performance. The assumptions of the method are supported by the analysis of a mathematical model, and its effectiveness is demonstrated by some experiments.