Issue No. 01 - January (1988 vol. 37)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/12.75149
An efficient vector-processing technique is proposed; it is based on a novel concept of multipipeline networking, which is generalized from the techniques of pipeline chaining and systolization. The authors also present the design principles of pipeline nets and provide programming, compiling and run-time techniques for converting scientific programs into pipeline net implementations. Performan
performance analysis; compilation; compound vector processing; vector-processing; multipipeline networking; pipeline chaining; systolization; pipeline nets; programming; run-time techniques; scientific programs; Livermore loops; parallel architectures; parallel programming; performance evaluation; pipeline processing; program compilers.
K. Hwang and Z. Xu, "Multipipeline Networking for Compound Vector Processing," in IEEE Transactions on Computers, vol. 37, no. , pp. 33-47, 1988.