ASP-DAC '06 Proceedings of the 2006 Asia and South Pacific Design Automation Conference
SBCCI '06 Proceedings of the 19th annual symposium on Integrated circuits and systems design
Proceedings of the 44th annual Design Automation Conference
An overview of reconfigurable hardware in embedded systems
EURASIP Journal on Embedded Systems
High-performance computing with desktop workstations
MATH'06 Proceedings of the 10th WSEAS International Conference on APPLIED MATHEMATICS
Reconfigurable Computing: The Theory and Practice of FPGA-Based Computation
Reconfigurable Computing: The Theory and Practice of FPGA-Based Computation
Modern development methods and tools for embedded reconfigurable systems: A survey
Integration, the VLSI Journal
International Journal of Reconfigurable Computing - Special issue on selected papers from ReConFig 2008
Bandwidth Management in Application Mapping for Dynamically Reconfigurable Architectures
ACM Transactions on Reconfigurable Technology and Systems (TRETS)
Concurrent calculations on reconfigurable logic devices applied to the analysis of video images
International Journal of Reconfigurable Computing - Special issue on selected papers from spl 2009 programmable logic and applications
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
A run-time partitioning algorithm for RTOS on reconfigurable hardware
EUC'05 Proceedings of the 2005 international conference on Embedded and Ubiquitous Computing
Elastic computing: A portable optimization framework for hybrid computers
Parallel Computing
| Hi-index | 0.00 |
The combination of hardware acceleration and flexibilitymake FPGAs important to image processing applications.There is also a need for efficient, flexible hardware/softwarecodesign environments that can balance the benefits andcosts of using FPGAs. Image processing applications oftenconsist of a pipeline of components where each componentapplies a different processing algorithm. Componentscan be implemented for FPGAs or software. Such systemsenable an image analyst to work with either FPGA or softwareimplementations of image processing algorithms for agiven problem. The pipeline assignment problem choosesfrom alternative implementations of pipeline componentsto yield the fastest pipeline. Our codesign system solvesthe pipeline assignment problem to provide the most effectiveimplementation automatically, so the image analyst canfocus solely on choosing components which make up thepipeline. However, the pipeline assignment problem is NPcomplete. An efficient, dynamic solution to the pipeline assignmentproblem is a desirable enabler of codesign systemswhich use both FPGA and software implementations.This paper is concerned with solving pipeline assignment inthis context. Consequently, we focus on optimal and heuristicmethods for fast (fixed time limit) runtime pipeline assignment.Exhaustive search, integer linear programmingand local search methods for pipeline assignment are investigated.We present experimental findings for pipelinesof 20 or fewer components which show that in our environment,optimal runtime solutions are possible for smallerpipelines and nearly optimal heuristic solutions are possiblefor larger pipelines.