This paper presents parallel algorithms for determining the number of partitions of a given integer N, where the partitions may be subject to restrictions, such as being composed of distinct parts, of a given number of parts, and/or of parts belonging to a specified set. We present a series of adapt
Partitioning and parallel algorithms for kidney models
โ Scribed by T.M. Hagstrom; R.P. Tewarson
- Publisher
- Elsevier Science
- Year
- 1988
- Tongue
- English
- Weight
- 243 KB
- Volume
- 11
- Category
- Article
- ISSN
- 0895-7177
No coin nor oath required. For personal study only.
โฆ Synopsis
A natural partitioning exists for kidney models. For example, if all the variables and equations associated with boundary values and interstitim are classified as basic, then the equations and variables corresponding to each tube can be solved independently. It will be shm how partitioning techniques of this type, used in conjunction with quasi-Newton methods, make it possible to handle -prehensive and realistic models of the kidney on small canputers and also leads to efficient utilization of parallel canputers. The developnent of parallel algorithms using general decanpositions arising in a variety of physical processes will also be discussed.
๐ SIMILAR VOLUMES
We introduce a class of layered graphs which we call \((k, 2)\) partite and which we argue are an interesting class because of several important applications. We show that testing for \((k, 2)\) partiteness can be done efficiently both on sequential and parallel machines, by showing that membership
A multiple-nephron inner medullary model is extended to include the outer medulla. The number of solutes modeled is also increased to include osmolyte in the central compartment. Two efficient parallel algorithms are described for solving the systems of differential and algebraic equations arising i
This paper presents a clustering algorithm that partitions node-labelled and edge-labelled ลฝ . directed acyclic precedence graphs APG into clusters such that all the clusters have balanced amount of computation load and there is only one communication path between any pair of clusters. The algorithm