Life after failure

This paper discusses the challenges, strategy and mitigation methods to prolong the operation of a product even after it is either about to have a failure or have experienced actual failures. The focus is mostly on two component types ASICs and memories. The aging and failure mitigation techniques vary by component type. Design for reliability techniques so far have been focused on measuring the impact of failures and how to prevent them as much as possible. The current advances in DFR methodologies are reaching their maximum level of effectiveness, still ASIC and memory failures are occurring impacting service operation of many products in the industry. Today´s technologies and their challenges require a different approach to addressing failures, which is the ability to have real time self-healing capability. The methodologies that have been developed in the area of self-healing have shown a promising future in the product overall fault management where failures due to aging ASICs and failing memories can be addressed while the system is in operation. This capability allows equipment providers to maintain their customer´s confidence in their ability to design, develop and deliver robust products and allow continuous system operation in the presence of failures. In this paper, we discuss the fundamental and phenomenon of aging, modeling and simulation, correlation between simulation and actual devices, sensor designs, mitigation strategies, impact on system reliability and availability and the reliability prediction.