Abstract

Efficient hardware/software partitioning is crucial towards realizing optimal solutions for constraint driven embedded systems. The size of the total solution space is typically quite large for this problem. In this paper, we show that the knapsack model could be employed for the rapid identification of hardware components that provide for time efficient implementations. In particular, we propose a method to split the problem into standard 0-1 knapsack problems in order to leverage on the classical approaches. The proposed method relies on the tight lower and upper bounds for each of these knapsack problems for the rapid elimination of the sub-problems, which are guaranteed not to give optimal results. Experimental results show that, for problem sizes ranging from 30 to 3000, the optimal solution of the whole problem can be obtained by solving only 1 sub-problem except for one case where it required the solution of 3 sub-problems.

Keywords

Cite this article

@article{cai-1455,
  author  = {Abhijit Ray and Wu Jigang and Thambipillai Srikanthan},
  title   = {Knapsack Model and Algorithm for Hardware/Software Partitioning Problem},
  journal = {Computing and Informatics},
  volume  = {23},
  number  = {5-6},
  pages   = {557--569},
  year    = {2004},
  issn    = {1335-9150},
  url     = {https://cai.type.sk/content/2004/5-6/knapsack-model-and-algorithm-for-hardware-software-partitioning-problem/}
}
Download .bib