IUCRC Planning Grant UC Davis: Center for Memory System Research (CMEMSYS)

点击翻译按钮获取中文摘要

Memory subsystem is a critical bottleneck that limits both performance and energy efficiency of artificial intelligence and data analysis workloads that are a cornerstone of many emerging applications with high societal impact. Recent advances in semiconductor and packaging technologies can enable new memory systems with significantly higher capacity, better performance and energy efficiency, and new computational paradigms such as processing in memory. To take advantage of these trends research must be carried out to integrate the new memory systems into the existing software and hardware ecosystems. Such research is critical to the industry because it enables businesses find new growth opportunities and recruit new talent.   CEMSYS aims to address the unmet industry research needs in integrating and evaluating emerging memory systems and evolving its applications to make the best usage of emerging memory technologies. CEMSYS will work in the following areas: (1) heterogeneous memory; (2) disaggregated memory; (3) big memory systems; (4) processing in memory; (5) memory reliability and security; (6) memory profiling and optimization; (7) emerging interconnects and packaging technologies, and (8) persistent memory.  Memory is the foundation of computer systems. The performance of almost all computer programs relies on memory for information storage and retrieval. Increasing capacity, performance, reliability, and energy efficiency of memory systems can benefit most applications. Hence, CEMSYS research has high economic significance with a broad societal impact. By establishing CEMSYS at UC Davis, a leading research university with strong commitment to diversity and equity, we bring unique research and development opportunities to the region. CEMSYS will mentor and develop a diverse, highly skilled, globally competitive engineering workforce. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.