CyberSP Seminars
March 15, 2019: Supercharge analytics over large IoT data: an OS approach
Dr. Felix Lin, Purdue University
Time and place: March 15, 2019, 10-11am, R1-101/101A
Abstract: This talk overviews our inquiry in systems software for taming large IoT data. I will describe two systems that target two crucial, complementary analytics paradigms: a data engine for processing large telemetry streams (hot data); a data store for querying large archival videos (cold data). The two systems exploit emerging hardware (e.g. 3D-stacked DRAM) as well as emerging workloads (e.g. neural networks). Both systems advance the state-of-the-art performance by orders of magnitude.
Our experiences highlight the significance of designing OSes for specific, important scenarios. The resultant OSes play key roles: mapping AI to new hardware, dynamically configuring AI, and trading off among competing objectives.
Bio: Felix Xiaozhu Lin is an assistant professor of Purdue ECE. He received PhD in CS from Rice and BS/MS from Tsinghua. At Purdue, he leads the Xroads systems exploration lab (XSEL) to accelerate and secure important computing scenarios. He and his students measure and build systems software with diverse techniques, including novel OS structures, kernel subsystem design, binary translation, and user-level runtimes. See http://xsel.rocks for more information. He is a recipient of a Google Faculty Award, an NSF CRII award, and the best paper award from ASPLOS’14.
Nov 9, 2018: Fuzzing and Securing the Server-Side Event-Driven Architecture
Dr. Dongyoon Lee, Virginia Tech
Abstract: The software development community is adopting the Event-Driven Architecture (EDA) to provide scalable web services, most prominently through Node.js. Though the EDA scales well, it comes with two inherent risks: concurrency errors and Event Handler Poisoning (EHP) Denial of Service attacks. Just as thread-based programs can have concurrency errors between unordered threads, event-driven programs may have them between unordered events. When an EDA-based server multiplexes many clients onto few threads, a blocked thread (EHP) renders the whole server unresponsive.
In this talk, I present Node.fz and Node.cure to address these problems. First, Node.fz provides a schedule fuzzing test tool that randomly perturbs the execution of a Node.js program, allowing Node.js developers to explore a variety of possible schedules during testing. Second, Node.cure proposes First-Class Timeouts, which incorporates timeouts at the EDA framework level, defending Node.js applications against all known EHP attacks.
Bio: Dongyoon Lee is an Assistant Professor in the Computer Science department at Virginia Tech. He obtained the M.S. (2009) and Ph.D. (2013) degrees in Computer Science and Engineering at the University of Michigan, Ann Arbor. Before joining Virginia Tech, he worked as an academic visitor in the next generation middleware platforms department at IBM T. J. Watson Research Center (Fall 2013). He also interned in the operating systems group at Microsoft Research, Redmond (Summer 2012), and in the systems analysis and verification department at NEC Laboratories America (Summer 2011). He received a Virginia Tech ICTAS Junior Faculty Award in 2017, a Google Research Award in 2015, a ProQuest Distinguished Dissertation Award in 2013, and a VMWare Graduate Fellowship in 2011. His co-authored papers won the best student paper finalist at SC 2016, and the best paper at ASPLOS 2011.