Desenvolvimento de um sensor "On-Chip" para monitoramento do envelhecimento de SRAMs

Abstract

Advances in Complementary Metal-Oxide Semiconductor (CMOS) technology have made possible the integration of millions of transistors into a small area, allowing the increase of circuits' density. In more detail, technology scaling caused the reduction of the transistors' delay, which has resulted in a signi cantly performance improvement of Integrated Circuits (ICs). Furthermore, the increase in the integration level of ICs allowed the development of ICs able to include an increasing number of functions, which in turn increased signi cantly their complexity. In parallel, the rapidly increasing need to store more information results in the fact that the Static Random Access Memory (SRAM) can occupy great part of the System-on-Chip (SoC) silicon area. This is con rmed by the SIA Roadmap which forecasts a memory density approaching 94% of the SoC area in about 10 years [1]. Consequently, memory has become the main responsible of the overall SoC area. However, the reduction of transistor size has introduced several reliability concerns that need to be a ronted by the adoption of di erent optimization techniques. In this context it is important to highlight the phenomenon known as Negative Bias Temperature Instability (NBTI), which a ects the reliability of the ICs along their lifes. Speci cally in the SRAMs NBTI causes degradation of the Static Noise Margim(SNM) which a ects the storage capacity of the memory cells. In this context, the main goal of this thesis is to specify, implement, validate and evaluate a hardware-based technique able to monitor the aging of SRAM cells in order to guarantee their reliability of during the lifetime. The proposed technique is based on an on-chip sensor capable of monitoring dynamic power consumption of the cells during write operations in order to compare them with the value set as default to a new cell. Finally, the proposed methodology has been functionally validated and its e ciency has been evaluated based on the analysis of its monitoring and detection capabilities and from the analysis of the introduced overheads as well as its immunity to the manufacturing process variation.