EDAC (Error Detection and Correction) techniques guarantee PVT variation safety by dynamically fixing timing error instead of providing static margins. However, previous EDAC works introduce additional area, power and performance penalty, thus the benefit from timing margin eliminating is limited. In this paper, we propose a novel EDAC Flip-Flop, EDSU, with ultra-low area overhead and nearly zero performance penalty. EDSU utilizes only two more transistors than conventional D-Flip Flop and can correct timing error simultaneously with detection. The ultra-lightweight property can obviously reduce area overhead and clock load, thus improve the variation tolerance ability and energy efficiency. EDSU is implemented in a commercial processor at SMIC 40 nm technology to evaluate its benefits. Simulation result shows EDSU inserted system gains 12.5% more performance at fixed voltage, 25% more variation tolerance and 10.5% energy saving at fixed throughput than state-of-art EDAC work.
Reducing circuit supply voltage to near-threshold (NT) region is an effective technique for achieving better energy efficiency in current ultra-low power circuit design. However, circuit variation problem becomes extremely worse in NT region and requires more design margin. Error detection and correction (EDAC) designs can remove circuit margin by solving transient error dynamically, but many of the traditional EDAC designs cause large area overhead and cycle per instruction (CPI) penalty. A compact timing error mask flip-flop (EMFF) based on error masking with ultra-low overhead by only adding six transistors (four for error detection and two for error correction) to the conventional flip-flop with 0 CPI penalty is presented. The proposed EMFF reduces system area overhead significantly. Therefore, the timing error tolerance ability is expanded and higher energy efficiency can be archived. EMFF is realised in an industrial processor in SMIC 40 nm CMOS with only 6.8% area overhead, compared with the original processor. At 0.5 V, the EMFF system gains 18.6% performance increasing than the most compact previous work with 3% higher efficiency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.