Communication efficient basic linear algebra computations on hypercube architectures
S. Lennart Johnsson · Journal of Parallel and Distributed Computing · 1987
This paper presents a few algorithms for embedding loops and multidimensional arrays in hypercubes with emphasis on proximity preserving embeddings. A proximity preserving embedding minimizes the need for communication bandwidth in computations requiring nearest neighbor communication. Two storage schemes for “large” problems on “small” machines are suggested and analyzed, and algorithms for matrix transpose, multiplying matrices, factoring matrices, and solving triangular linear systems are presented. A few complete binary tree embeddings are described and analyzed. The data movement in the matrix algorithms is analyzed and it is shown that in the majority of cases the directed routing paths intersect only at nodes of the hypercube allowing for a maximum degree of pipelining.
Cited by 239 works. doi:10.1016/0743-7315(87)90002-5
Authors
References
- Parallel Processing with the Perfect Shuffle
- The cosmic cube
- Very high-speed computing systems
- The NYU Ultracomputer—Designing an MIMD Shared Memory Parallel Computer
- Nested Dissection of a Regular Finite Element Mesh
- A Parallel Algorithm for the Efficient Solution of a General Class of Recurrence Equations
- Topological properties of hypercubes
- Ultracomputers
- Implementing Linear Algebra Algorithms for Dense Matrices on a Vector Pipeline Machine
- A Survey of Parallel Machine Organization and Programming