๐”– Bobbio Scriptorium
โœฆ   LIBER   โœฆ

Program Speedup in a Heterogeneous Computing Network

โœ Scribed by V. Donaldson; F. Berman; R. Paturi


Publisher
Elsevier Science
Year
1994
Tongue
English
Weight
597 KB
Volume
21
Category
Article
ISSN
0743-7315

No coin nor oath required. For personal study only.

โœฆ Synopsis


Program speedup is an important measure of the performance of an algorithm on a parallel machine. Of particular importance is the near linear or superlinear speedup exhibited by the most performance-efficient algorithms for a given system. We describe network and program models for heterogeneous networks, define notions of speedup and superlinear speedup, and observe that speedup consists of both heterogeneous and parallel components. We also consider the case of linear tasks, give a lower bound for the speedup, and show that there is no theoretical upper limit on heterogeneous speedup. O 1994 Academic Press, Inc.


๐Ÿ“œ SIMILAR VOLUMES


Predicting Speedup for Distributed Compu
โœ Phillip M. Feldman; Raisa E. Feldman; David B. Kim ๐Ÿ“‚ Article ๐Ÿ“… 1997 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 145 KB

Before a conventional application is converted into a distributed one (typically a costly process), it is prudent to estimate the improvement in run time that will be achieved. Previous research has tended to ignore communications delays in order to facilitate analysis. However, such models lead to

Determining the Execution Time Distribut
โœ Yan Alexander Li; John K. Antonio; Howard Jay Siegel; Min Tan; Daniel W. Watson ๐Ÿ“‚ Article ๐Ÿ“… 1997 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 278 KB

An important problem in heterogeneous computing (HC) is predicting task execution time. A methodology is introduced for determining the execution time distribution for a given data parallel program that is to be executed in an SIMD, MIMD (SPMD), and/or mixed-mode SIMD/MIMD (SPMD) HC environment. The

A parallel language and its programming
โœ Lastovetsky, Alexey ;Arapov, Dmitry ;Kalinov, Alexey ;Ledovskih, Ilya ๐Ÿ“‚ Article ๐Ÿ“… 2000 ๐Ÿ› John Wiley and Sons ๐ŸŒ English โš– 201 KB ๐Ÿ‘ 1 views

The paper presents a new parallel language, mpC, designed specially for programming high-performance computations on heterogeneous networks of computers, as well as its supportive programming environment. The main idea underlying mpC is that an mpC application explicitly defines an abstract network

A computer program for simulating a netw
โœ Donald H. Perkel ๐Ÿ“‚ Article ๐Ÿ“… 1976 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 490 KB

Describing a neural network to be simulated by PABLO and controlling its operation are accomplished through a variety of control cards. With a few exceptions, these have a standard format. Each type of card has its particular function, e.g., starting a new network. defining network components by spe