Ž . Parallel Computing 24 1998 221–245 A method for exploiting communicationrcomputation overlap in hypercubes 1 Luis Dıaz de Cerio ) , Miguel Valero-Garcıa 2 , Antonio Gonzalez 3 ´ ´ ´ Departament d’Arquitectura de Computadors, UniÕersitat Politecnica de Catalunya, c r Jordi Girona 1– 3, ` Campus Nord — Edifici D6, E-08034 Barcelona, Spain Received 19 December 1996; revised 29 July 1997 Abstract 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 distinguish- ing 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. q 1998 Elsevier Science B.V. All rights reserved. Keywords: Hypercube; Communication pipelining; Communicationrcomputation overlap; Performance mod- eling; FFT; Vector Add ) Corresponding author. E-mail: ldiaz@ac.upc.es 1 Ž . Expanded version of a talk presented at Euro-Par’96 Lyon, France, August 1996 . 2 E-mail: miguel@ac.upc.es 3 E-mail: antonio@ac.upc.es 0167-8191r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. Ž . PII S0167-8191 98 00005-2