Application-specific programmable control for high-performance asynchronous circuits

Citation
Hm. Jacobson et G. Gopalakrishnan, Application-specific programmable control for high-performance asynchronous circuits, P IEEE, 87(2), 1999, pp. 319-331
Citations number
24
Categorie Soggetti
Eletrical & Eletronics Engineeing
Journal title
PROCEEDINGS OF THE IEEE
ISSN journal
00189219 → ACNP
Volume
87
Issue
2
Year of publication
1999
Pages
319 - 331
Database
ISI
SICI code
0018-9219(199902)87:2<319:APCFHA>2.0.ZU;2-P
Abstract
The advantages of the programmable control paradigm are widely known in the design of synchronous sequential circuits: easy correction of late design errors, easy upgrade of product families to meet time-to-market constraints , and modifications of the control algorithm, even at run time. However, de spite the growing interest in asynchronous (self-timed) circuits, programma ble asynchronous controllers based on the idea of microprogramming have not been actively pursued In this paper, we propose an asynchronous microprogr ammed control organization (called a microengine) that targets application- specific implementations and emphasizes simplicity, modularity, and high pe rformance. The architecture takes advantage of the natural ability of self- timed circuits to chain actions efficiently without the clock-based schedul ing constraints that would be involved in comparable synchronous designs. T he result is a general approach to the design of application-specific micro engines featuring a programmable data-path topology that offers very compac t microcode and high performance-in fact, performance close to that offered by automated high-level synthesis tools targeting stare-of-the-art asynchr onous hardwired controllers. In performance comparisons of a CD-player erro r decoder design, the proposed microengine architecture was 26 times faster than the general purpose hardware of a 280 MIPS microprocessor, over three times as fast as the special purpose hardware of a low-power macromodule b ased implementation, and even slightly faster than a finite state machine-b ased implementation.