Many modern integrated circuits are sequenced based on globally distributed periodic timing signals called clocks. This method of sequencing, synchronous, is prevalent and has contributed to the remarkable advancements in the semiconductor industry in form of chip density and speed in the last decades. For the trend to continue as proposed in Moore’s law, the number of transistors on a chip doubles about every two years, there are increasing requirements for enormous circuit complexity and transistor downscaling.
As the industry pursues these factors, many problems associated with switching delay, complexity management and clock distribution have placed limitation on the performance of synchronous system with an acceptable level of reliability. Consequently, the synchronous system design is challenged on foreseeable progress in device technology.
These concerns and other factors have caused resurgence in interest in the design of asynchronous or self-timed circuits that achieve sequencing without global clocks. Instead, synchronization among circuit elements is achieved through local handshakes based on generation and detection of request and acknowledgement signals.
Tekedia Mini-MBA edition 16 (Feb 10 – May 3, 2025) opens registrations; register today for early bird discounts.
Tekedia AI in Business Masterclass opens registrations here.
Join Tekedia Capital Syndicate and invest in Africa’s finest startups here.
Some notable advantages of asynchronous circuits over their synchronous counterparts are presented below:
* Average case performance. Synchronous circuits have to wait until all possible computations have completed before producing the results, thereby yielding the worst-case performance. In the asynchronous circuits, the system senses when computation has completed thereby enabling average case performance. For circuits like ripple carry adders with significantly worst-case delay than average-case delay, this can be an enormous saving in time.
* Design flexibility and cost reduction, with higher level logic design separated from lower timing design
* Separation of timing from functional correctness in certain types of asynchronous design styles thereby enabling insensitivity to delay variance in layout design, fabrication process, and operating environments.
* The asynchronous circuits consume less power than synchronous since signal transitions occur only in areas involved in current computation.
* The problem of clock skew evident in synchronous circuit is eliminated in the asynchronous circuit since there is no global clock to distribute. The clock skew, difference in arrival times of clock signal at different parts of the circuit, is one of the major problems in the synchronous design as feature size of transistors continues to decrease.
author/ndubuisi ekekwe