On the notion of granularity
β Scribed by Clyde P. Kruskal; Carl H. Smith
- Publisher
- Springer US
- Year
- 1988
- Tongue
- English
- Weight
- 658 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0920-8542
No coin nor oath required. For personal study only.
β¦ Synopsis
Granularity is a well known concept in parallel processing. While intuitively, the distinction between coarse-grain and fine-grain paraMlism is clear, there is no rigorous definition. This paper develops two notions of granuIarity, each defined formally and represented by a single rational number. The two notions are compared and contrasted with each other and with previously proposed definitions of granularity.
π SIMILAR VOLUMES
In this paper we introduce a correct notion of fuzzy consensus that is a true generalization of the well-known Boolean consensus result. Our definition of fuzzy consensus becomes identical to the Boolean consensus when the fuzzy variables are allowed to take only Boolean values. It is also shown tha
## Abstract Based on the terms βendβ and βcofinal spanning subtreeβ a general notion of Hamiltonicity of infinite graphs is developed. It is shown that the cube of every connected locally finite graph is Hamiltonian in this generalized sense.