Speak: Dr. Liang Shi
Time: 14:00-15:00, Oct. 31
Host: Dr. Shu Yin
Flash memory has been widely used in computer systems including embedded systems, personal computers, and data centers etc., for its high speed and low power consumption. Different from traditional storage devices, flash memory has some unique specifications such as black-box design, which provide users an traditional HDD API abstractions; non-in-situ data updates, where data is always written as new and requires garbage collection mechanisms for capacity; hierarchical parallelism structures, which requires managements to fully take the advantages of multi-level parallelism. In this talk, we first discuss the fundamental characteristics of flash memory devices organization. We then introduce hardware-software codesigns from system layers, controller layers, and device layers. Our proposed solution provides better hardware and software compatibility for flash storage systems and improves I/O performance, reliability and lifespan of flash devices.
Dr. Liang Shi received Ph.D degree from Department of Computer Science at the University of Science and Technology of China, in 2013. He also received the joint PhD degrees from Department of Computer Science at the City University of Hong Kong. He received his B.E degrees in Computer Science from University of Xi'an Post and Telecommunication, Xi'an, Shanxi, China, in 2008, respectively. Liang Shi is currently a professor at East China Normal University. Before that, he was an associate professor in the College of Computer Science, Chongqing University. In addition, he also was a postdoc at City University of Hong Kong from Mar. 2015 to Jul. 2016.
Dr. Shi's research interests include storage systems, embedded systems, and non-volatile memories (NVM). Liang Shi has published more than 70 papers, including over 30 IEEE/ACM Transactions papers, and many top conference papers in the fields of embedded systems and EDA, e.g., FAST, MSST, DAC, DATE, LCTES, ISLPED, etc. Liang Shi won the Best Paper Award at NVMSA 2015, and the best paper award nominee at ASP-DAC 2017. The H-index is 14.
Sist seminar 18215