The hyperplane method for an array computer
Techniques are described for implementing the hyperplane method for an array computer with a limited form of memory-processor interconnection, such as the Illiac IV. A new pipelined hyperplane method is introduced. It is useful for compiling a convergent iterative computation, if the computation is programmed with a non-deterministic end of loop test.
Unable to display preview. Download preview PDF.
- L. Lamport, "The Parallel Execution of DO Loops," Comm. ACM 17, 2 (Feb., 1974), pp. 83–93.Google Scholar
- D.E. McIntyre, "An Introduction to the Illiac IV Computer," Datamation, 16, 4 (April, 1970), pp. 60–67.Google Scholar
- R.E. Millstein, "Control Structures in Illiac IV Fortran," Comm. ACM 16, 10 (October, 1973), pp. 621–627.Google Scholar