Browse Prior Art Database

Performance Enhancement with PLA Interleaving

IP.com Disclosure Number: IPCOM000049852D
Original Publication Date: 1982-Aug-01
Included in the Prior Art Database: 2005-Feb-09
Document File: 2 page(s) / 42K

Publishing Venue

IBM

Related People

Leininger, JC: AUTHOR [+3]

Abstract

Programmable logic array (PLA) performance is determined by the technology, array sizes, power dissipation, and Physical size. When desired Performance is more than the actual PLA performance, PLAs may be interleaved providing the inputs can be sent independently to each PLA. The additional overhead associated with multiplexing is duplication of the input decoder and one additional input to each output latch.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 100% of the total text.

Page 1 of 2

Performance Enhancement with PLA Interleaving

Programmable logic array (PLA) performance is determined by the technology, array sizes, power dissipation, and Physical size.

When desired Performance is more than the actual PLA performance, PLAs may be interleaved providing the inputs can be sent independently to each PLA. The additional overhead associated with multiplexing is duplication of the input decoder and one additional input to each output latch.

A PLA has an input decode, AND array, OR array and output latches. If the inputs are changed each cycle and the outputs are latched at the end of each cycle, then the delay through the PLA determines the minimum cycle time that can be achieved. In some cases, the delay through one large PLA can determine the overall system cycle time and system performance limit. As PLAs increase in size, the delay time correspondingly increases. If the PLA inputs can be generated in pairs, with one set containing the data for cycles N, N+2, N+4, etc., and the other set containing data for cycles N+1, N+3, N+5, etc., then, as shown in the figure, the PLAs may be interleaved and the physical size of each reduced. PLA latches will be set alternately with data from PLA #1 or PLA #2. Except for the first cycle, the effective delay will be one half of that obtained when the PLA was not divided into two parts.

1

Page 2 of 2

2

[This page contains 4 pictures or other non-text objects]