Memory designs require timing margins to compensate for aging and fabrication process variations. With technology downscaling, aging mechanisms became more apparent, and larger margins are considered necessary. This, in return, means a larger area requirement and lower performance for the memory. Bias Temperature Instability (BTI) is one of the main contributors to aging, which slows down transistors and ultimately causes permanent faults. In this paper, first, we propose a low-cost aging mitigation scheme, which can be applied to existing hardware to mitigate aging on memory address decoder logic. We mitigate the BTI effect on critical transistors by applying a rejuvenation workload to the memory. Such an auxiliary workload is executed periodically to rejuvenate transistors that are located on critical paths of the address decoder. Second, we analyze workloads' efficiency to optimize the mitigation scheme. Experimental results performed with realistic benchmarks demonstrate several-times lifetime extension with a negligible execution overhead.@en

The Soft-Error (SE) reliability and the effects of Negative Bias Temperature Instability (NBTI) in deep submicron technologies are characterized as the major critical issues of high-performance integrated circuits. The previous scientific research studies provide a comprehensive description that the soft-error vulnerability becomes more severe as the circuit performance degrades with aging. The main reason is the reduction of cell-level critical charge in an aging environment. However, such increased soft-error generation does not necessarily contribute towards circuits' critical functional failures. The proposal of this paper is the experimental investigation of soft error propagation at the aged gate-level by considering the different derating factors like Electrical Derating (EDR), Temporal Derating (TDR), Logical Derating (LDR), and Functional Derating (FDR). As contrary to the previous studies, the results of this work prove that SEU fault propagation probability is reducing in critical paths as time advances while the propagation probability of SET faults is neither reducing nor increasing, but the spot of generation of failure enhancing SETs is shifting within the clock period.@en

Manufacturing defects can cause faults in FinFET SRAMs. Of them, easy-to-detect (ETD) faults always cause incorrect behavior, and therefore are easily detected by applying sequences of write and read operations. However, hard-to-detect (HTD) faults may not cause incorrect behavior, only parametric deviations. Detection of these faults is of major importance as they may lead to test escapes. This paper proposes a new design-for-testability (DFT) scheme for FinFET SRAMs to detect such faults by creating a mismatch in the sense amplifier (SA). This mismatch, combined with the defect in the cell, will incorrectly bias the SA and cause incorrect read outputs. Furthermore, post-silicon calibration schemes can be used to avoid over-testing or test escapes caused by process variation effects. Compared to the state of the art, this scheme introduces negligible overheads in area and test time while it significantly improves fault coverage and reduces the number of test escapes.

@en