Jun Wang will present a Computer Science Colloquium virtual lecture on Friday, October 21, 2022, from 3-4 p.m. Wang is a professor of computer engineering and director of the Computer Architecture and Storage Systems (CASS) Laboratory at the University of Central Florida, Orlando.
Dr. Wang has conducted extensive research in the areas of computer systems and data-intensive computing. He is a Fellow of IEEE.
The lecture is presented by the Michigan Tech Department of Computer Science.
Talk Title: “High-performance Data Processing via Hardware/Software Co-design and Sampling”
Talk Abstract: In today’s big data and big compute era, the explosive rate of data growth strangles the limited scalability of the DRAM technology, which defies the performance potentials for in-memory applications. Fortunately, emerging ultra-low latency and terabyte scale non-volatile memory (NVM) technologies, such as 3D-Xpoint, Phase-Change Memory (PCM) and Memristor, are promising candidates for supplementing or even replacing DRAM. Emerging NVMs are very dense, hence promise large capacities. Additionally, NVMs are non-volatile, thus enable persistent applications and byte-addressable files. Both density and persistency are key enablers for in-memory applications.
On the other side, emerging NVMs are slower than DRAM, constrained by endurance and remanence problems, thus revamping and optimizing current OS with architecture support for finer-granularity, locality, and avoiding contentions are key aspects to unlock the NVM performance. Dr. Wang leads his lab to handle these challenges through several research thrusts. First, we develop new fine-granularity Copy-On-Write operations for Secure Non-Volatile Memories to tame the huge memory page problems for memory bulk operations. Second, we develop an architecture-aware sampling library to enable efficient and high-performing sampling through employing its knowledge of the NVM architectural details to maximize data locality and avoiding inter-thread contentions. Last, we develop a distribution-aware method to enable both efficient and accurate approximations on arbitrary sub-datasets of a large dataset.
Speaker Bio: Dr. Jun Wang is a Professor of Computer Engineering; and Director of the Computer Architecture and Storage Systems (CASS) Laboratory at the University of Central Florida, Orlando, FL, USA. He has conducted extensive research in the areas of Computer Systems and Data-Intensive Computing. His specific research interests include massive storage and file Systems in a local, distributed, and parallel systems environment. He is the recipient of the National Science Foundation Early Career Award 2009 and the Department of Energy Early Career Principal Investigator Award 2005. He has authored over 150 publications in premier journals and conferences. Dr. Wang is the Fellow of IEEE.