Cadenas, Oswaldo, Megson, Graham M. and Jones, David J. (2005) Implementation of a block based neural branch predictor. In: 8th Euromicro Conference on Digital System Design: proceedings. IEEE, pp. 235-238. ISBN 0769524338Full text not available from this repository.
This paper contributes to a dynamic branch predictor algorithm based on a perceptron in two directions: Firstly, a new block form of computation is introduced that reduces theoretically by half the combinational critical path for computing a prediction. Secondly, implementation in FPGA hardware is fully developed for quantitative comparison purposes. FPGA circuits for a one-cycle block predictor produces 1.7 faster clock rates than a direct implementation of the original perceptron predictor. This faster clock allows to realize predictions with longer history lengths for the same hardware budget.
|Item Type:||Book Section|
|Uncontrolled Keywords:||Combinational circuits, field programmable gate arrays, perceptrons, FPGA hardware, block based neural branch predictor, dynamic branch predictor algorithm, perceptron predictor|
|Subjects:||University of Westminster > Science and Technology > Electronics and Computer Science, School of (No longer in use)|
|Depositing User:||Miss Nina Watts|
|Date Deposited:||27 Jan 2009 15:28|
|Last Modified:||19 Oct 2009 13:15|
Actions (login required)
|Edit Item (Repository staff only)|