Optimized design verification methodology of variation-tolerant Nanoscale systems

点击翻译按钮获取中文摘要

Integrated circuit (IC) technology has improved dramatically in performance due to continued miniaturization of the transistors and wiring. However, as the minimum feature size shrinks below 32 nm, it becomes very difficult to ensure that all defect-free fabricated chips will work correctly despite unavoidable variations in the process parameters, voltage conditions and operating temperature (that is, PVT variations). The Canadian start-up Solido Design Automation Inc. has developed industry-leading design software that allows engineers to efficiently select, from many thousands of PVT scenarios, the worst-case conditions that need to be verified in simulation. An initial objective of this research project is to establish a mutually beneficial collaboration between University of Alberta researchers and Solido to investigate improved methods for verifying the correct operation of ICs in the presence of PVT variations. The ultimate original objective of the project is to develop a comprehensive design methodology supporting the formation of new effective tools aimed at the identification of worst-case PVT corners and a design of a suite of algorithms as well as circuit benchmarks. As a long-term strategic goal, we will form a sustainable, world class research center focused on innovative optimized design of PVT variation-oriented verification methodology, by coordinating with world class researchers and bring in other Canadian companies. The comprehensive expertise of the research team is multifaceted and fully covers the needs arising from the proposed research project. The developed methodology is not only suitable for the design of current systems but it is far fetching and goes beyond the boundaries of the current components. In emerging nonconventional nanotechnologies, nondeterministic behaviours become profoundly visible and associated with quantum effects, environmental noise and, in some cases, inexpensive but inaccurate molecular self-assembly. Therefore, variation resilience must be incorporated into any nanoscale system design. In this sense the proposed methodologies will be applicable to the design of emerging ICT systems and will benefit the Canadian industry in a long term.