Jonathan C. Hardwick, "Practical Parallel Divide-and-Conquer Algorithms", Ph.D. Thesis, School of Computer Science, Carnegie Mellon University, December 1997, 154 pp. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Design and Evaluation of the Hamal Parallel Computer - Grossman (2002) (1 citation) (Correct)
....specifying the processor nodes which should be used to operate on these structures. The permissible operations on processor sets are union, intersection, and split (split the set into two smaller sets according to a supplied ratio) Processor sets are nearly identical to processor teams in [Hardwick97] and spans in [Brown02a] the only difference is that both teams and spans are restricted to processor sets of the form [a, b] where the processors are numbered from 0 to N 1) Removing this restriction allows one to take the union of two processor sets and also provides the opportunity to ....
.... are numbered from 0 to N 1) Removing this restriction allows one to take the union of two processor sets and also provides the opportunity to allocate processor sets which reflect the physical layout of the nodes (e.g. create a processor set for a subcube of the nodes in a 3D mesh) In [Hardwick97] it is shown that the processor team abstraction provides simple and efficient support for divide and conquer algorithms. struct Vector int32 len; number of elements int32 elsize; size of elements ProcSet procs; Veclet veclets; sparse pointer ; struct Veclet int32 len; ....
Jonathan C. Hardwick, "Practical Parallel Divide-and-Conquer Algorithms", Ph.D. Thesis, School of Computer Science, Carnegie Mellon University, December 1997, 154 pp.
Implementation and Evaluation of an Efficient Parallel Delaunay.. - Hardwick (1997) (3 citations) Self-citation (Hardwick) (Correct)
....program which uses the algorithm by Blelloch et al. as a coarse parallel partitioner, switching to an efficient implementation of Dwyer s serial algorithm provided by the Triangle package [33] at the leaves of the recursion tree. The program was parallelized using the Machiavelli toolkit [24], Inner Convex Hull Outer Delaunay Triangulation Figure 1: Nested recursion in Delaunay triangulation algorithm by Blelloch et al. [8] Each recursive level of the outer divideand conquer triangulation algorithm, which has a perfect split, uses as a substep a divide and conquer convex hull ....
.... layer assumes a vector PRAM model [6] This can be efficiently implemented on vector processors with high memory bandwidth, but it is harder to do so on current RISC based NUMA multiprocessor architectures, due to the higher relative costs of communication and poor data locality [21] Machiavelli [24] is a new parallel toolkit for divide andconquer algorithms that is intended to alleviate some of these problems. It is designed to be usable both as an implementation layer for languages such as Nesl, and as a programmer s toolkit for the direct implementation of efficient parallel programs. ....
Jonathan C. Hardwick. Practical Parallel Divide-andConquer Algorithms. PhD thesis, School of Computer Science, Carnegie Mellon University, 1997. To appear.
Implementation and Evaluation of an Efficient 2D Parallel.. - Hardwick (1997) Self-citation (Hardwick) (Correct)
....program which uses the algorithm by Blelloch et al. as a coarse parallel partitioner, switching to an efficient implementation of Dwyer s serial algorithm provided by the Triangle package [33] at the leaves of the recursion tree. The program was parallelized using the Machiavelli toolkit [24], which has been designed both for the direct implementation of parallel divide and conquer algorithms (as in this case) and as an implementation layer for nested data parallel languages. It is particularly well suited to exploiting the nested divide and conquer nature of the algorithm by ....
.... layer assumes a vector PRAM model [6] This can be efficiently implemented on vector processors with high memory bandwidth, but it is harder to do so on current RISC based NUMA multiprocessor architectures, due to the higher relative costs of communication and poor data locality [21] Machiavelli [24] is a new parallel toolkit for divide and conquer algorithms that is intended to alleviate some of these problems. It is designed to be usable both as an implementation layer for languages such as Nesl, and as a programmer s toolkit for the direct implementation of efficient parallel programs. ....
Jonathan C. Hardwick. Practical Parallel Divide-and-Conquer Algorithms. PhD thesis, School of Computer Science, Carnegie Mellon University, 1997. To appear.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC