𝔖 Bobbio Scriptorium
✦   LIBER   ✦

A method for exploiting communication/computation overlap in hypercubes

✍ Scribed by Luis Dı́az de Cerio; Miguel Valero-Garcı́a; Antonio González

Publisher
Elsevier Science
Year
1998
Tongue
English
Weight
361 KB
Volume
24
Category
Article
ISSN
0167-8191

No coin nor oath required. For personal study only.

✦ Synopsis

This paper presents a method to derive efficient algorithms for hypercubes. The method . exploits two features of the underlying hardware: a the parallelism provided by the multiple . communication links of each node and b the possibility of overlapping computations and communications which is a feature of machines supporting an asynchronous communication protocol. The method can be applied to a generic class of hypercube algorithms whose distinguishing features are quite frequent in common algorithms for hypercubes. Many examples of this class of algorithms are found in the literature for different problems. The paper shows the efficiency of the method for two case studies. The results show that the reduction in communication overhead is very significant in many cases. They also show that the algorithms produced by our method are always very close to the optimum in terms of execution time.

📜 SIMILAR VOLUMES

A technique for overlapping computation
A technique for overlapping computation and communication for block recursive algorithms
✍ GUPTA, S. K. S.; HUANG, C.-H.; SADAYAPPAN, P.; JOHNSON, R. W. 📂 Article 📅 1998 🏛 John Wiley and Sons 🌐 English ⚖ 175 KB

This paper presents a design methodology for developing efficient distributed-memory parallel programs for block recursive algorithms such as the fast Fourier transform (FFT) and bitonic sort. This design methodology is specifically suited for most modern supercomputers having a distributed-memory a

A pipelined schedule to minimize complet
A pipelined schedule to minimize completion time for loop tiling with computation and communication overlapping
✍ N. Koziris; A. Sotiropoulos; G. Goumas 📂 Article 📅 2003 🏛 Elsevier Science 🌐 English ⚖ 292 KB

This paper proposes a new method for the problem of minimizing the execution time of nested for-loops using a tiling transformation. In our approach, we are interested not only in tile size and shape according to the required communication to computation ratio, but also in overall completion time. W

A computation method for induction motor
A computation method for induction motors in steady-state
✍ B. Laporte; G. Vinsard; J.C. Mercier 📂 Article 📅 1995 🏛 Elsevier Science 🌐 English ⚖ 370 KB

A model corresponding to a fast algorithm to compute the steady-state of the induction motor is presented. The principle of the model is to start from the idea that the time pulsation is to in the stator region and sto in the rotor region. The coupling between the rotor and the stator is made only b

A general computer simulation method for
A general computer simulation method for a 3-D received signal level in wideband mobile communications
✍ Xiongwen Zhao; Ioannis T. Rekanos; Pertti Vainikainen 📂 Article 📅 2001 🏛 John Wiley and Sons 🌐 English ⚖ 144 KB

## Abstract A wideband 3‐D received signal level has been simulated by assuming that the multipath number follows a Poisson distribution, the amplitudes follow either Rayleigh or Rice distributions, and the excess delay and the angle of arrival in the elevation plane follow exponential and Gaussian