Integer and floating point matrix-vector multiplication on the reconfigurable mesh
Document Type
Conference Proceeding
Publication Date
1-1-1996
Abstract
This paper presents the following results for matrix-vector multiplication on the reconfigurable mesh (R-Mesh). Multiplication of an N × N matrix and an N × 1 vector (each element of which is an integer of w bits) can be performed in O(log w) time on a two-dimensional O(N w/log w log2 N) × O(N w/log w log2 N) R-Mesh or a three-dimensional N × N × (w/log w log2 N) R-Mesh; in both cases, inputs and outputs are stored as w-bit integers. A natural extension of this problem is for floating point inputs; floating point numbers have not been handled before on any reconfigurable bus-based model. Matrix-vector multiplication with floating point inputs can be performed in O(max{log log N, log w}) time on an N × N × max {√N/log log N, w/log w log2 N} three-dimensional R-Mesh, with the inputs and outputs stored as w-bit floating point numbers.
Publication Source (Journal or Book title)
IEEE Symposium on Parallel and Distributed Processing - Proceedings
First Page
702
Last Page
706
Recommended Citation
Trahan, J., Lu, C., & Vaidyanathan, R. (1996). Integer and floating point matrix-vector multiplication on the reconfigurable mesh. IEEE Symposium on Parallel and Distributed Processing - Proceedings, 702-706. Retrieved from https://repository.lsu.edu/eecs_pubs/1758