The connection machine
The LDF 100: a large grain dataflow parallel processor
ACM SIGARCH Computer Architecture News
An Orthogonal Multiprocessor for Parallel Scientific Computations
IEEE Transactions on Computers
The structure of parallel computation and computers for graphics procedures
The structure of parallel computation and computers for graphics procedures
Your Favorite Parallel Algorithms Might Not Be as Fast as You Think
IEEE Transactions on Computers
A taxonomy of scheduling in general-purpose distributed computing systems
IEEE Transactions on Software Engineering
Hi-index | 0.00 |
Procedure-based communication can convert variant communication patterns of parallel computation to a simple data sending and receiving process. This paper describes a general purpose, MIMD parallel architecture that effectively supports the procedure-based communication. The system uses a two-dimensional memory array and a multiple bus system as data transfer media. The system eliminates bus contention and reduces memory access conflicts. In spite of its partially shared memory, the architecture guarantees direct data access for all processors if a procedure-based communication is employed.