| M. Amamiya, "Data Flow Computing and Parallel Reduction Machine," Future Generation Computer Systems, Vol.4, No.1, pp.53-67 (1988). |
....ffl Pure fine grain dataflow execution scheme employs a by value data access mechanism, and this causes explicit data copying. This explicit data copying is redundant, and should be eliminated by introducing by reference data access in which data are accessed only when they are needed [1]. ffl Matching memory mechanism requires complex and expensive hardware. This also makes it difficult to use caches. Furthermore, since dataflow machines are implemented as circular pipeline, they have the following problems: ffl Dataflow machines are ineffective when they execute sequential ....
M. Amamiya, "Data Flow Computing and Parallel Reduction Machine," Future Generation Computer Systems, Vol.4, No.1, pp.53-67 (1988).
....ffl Pure fine grain dataflow execution scheme employs a by value data access mechanism, and this causes explicit data copying. This explicit data copying is redundant, and should be eliminated by introducing by reference data access in which data are accessed only when they are needed [2]. ffl Matching memory mechanism requires complex and expensive hardware. This also makes it difficult to use caches. Furthermore, since dataflow machines are implemented as circular pipeline, they have the following problems: ffl Dataflow machines are ineffective when they execute sequential ....
M. Amamiya, "Data Flow Computing and Parallel Reduction Machine," Future Generation Computer Systems, Vol.4, No.1, pp.53-67 (1988).
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