Computing with Resistive Memory Technologies

作者: 分类: 学术报告 时间: 2015-12-31 评论: 暂无评论

题目:Computing with Resistive Memory Technologies

摘要:For decades, technology scaling has made it possible to design systems with faster processors and larger memory systems. As technology scales below the 14 nm node, however, conventional CMOS scaling is facing fundamental physical limits. With more than two billion transistors integrated on a single die, processor power dissipation now exhausts the capability of conventional cooling technologies. On the memory side, DRAM density scaling has become increasingly difficult due to the challenges in maintaining a sufficiently high storage capacitance and a sufficiently low leakage current at nanoscale feature sizes. In response, industry has started investing in emerging resistive memory technologies (e.g., STT-MRAM, PCM, and RRAM), which hold the potential to replace memories based on SRAM, DRAM, and Flash. This talk will examine our work on leveraging resistive memory technologies in designing energy-efficient microprocessor cores, and novel accelerators that enable computer systems with qualitatively new capabilities.

报告人:Dr. Xiaochen Guo is an Assistant Professor at Lehigh University. She received her Ph.D. degree from the University of Rochester. Her research leverages resistive memories to build energy-efficient processors, memory systems, and accelerators. She was awarded the IBM Ph.D. Fellowship twice; and interned at Samsung Research America and IBM T. J. Watson Research Center.

标签: none