I work at Clockwork Systems as Staff Research Engineer. I obtained my PhD degree in Dept. of Electrical Engineering at Stanford University, working with Prof. Balaji Prabhakar, Prof. Mendel Rosenblum, Prof. Mohammad Alizadeh, and Prof. Anirudh Sivaraman. I was affliated to Stanford Platform Lab. I got my Bachelor degree in Electronic Engineering from Tsinghua University.

My research interests include computer networking and distributed systems.

Email: liushiyu1993 [at] gmail [dot] com.

Google Scholar

LinkedIn

At Clockwork Systems, I lead the team who is responsible for R&D of Product Suite Cloud Deluxe which commercializes my PhD work of On-Ramp and our patents. It addresses several key technical challenges in network and cloud infra that impede performance and drive-up costs for customer-facing applications and AI/ML workloads. The key products include (read the testimonials by executives of tech giants):

• Packet Rocket: Detect and mitigate network congestion. Intelligently pace packets to kill network latencies and eliminate packet drops for optimal application performance.
• Ghost Buffer: Reserve bandwidth headroom to achieve near-minimum network latency for delay-sensitive applications.
• Bandwidth Slicer: Create programmable bandwidth slices between VMs. Prioritize traffic and ensure critical applications have the necessary bandwidth.

At Stanford, my research topics included edge-based sensing, inferring, learning and controlling in data center networks, e.g. clock synchronization (Huygens), network telemetry and monitoring (SIMON), flow control (On-Ramp) in both on-prem data centers and public clouds, as well as building highly scalable and fault-tolerant distributed systems (e.g. ClockChain, CloudEx) on top of them. Check the following papers and my talks in Stanford Platform Lab annual reviews and retreats for more details.

My research has drawn wide interests from industry. The works on clock sync, network telemetry, and flow control are being commercialized at Clockwork Systems Inc. They have also been piloted or adopted in major technology companies (e.g. Google, eBay, Facebook etc.) and financial institutions (e.g. stock exchanges like Nasdaq, hedge funds, banks). Also check the media coverage for details.

• NSDI '21 "Breaking the Transience-Equilibrium Nexus: A New Approach to Datacenter Packet Transport", S. Liu, A. Ghalayini, M. Alizadeh, B. Prabhakar, M. Rosenblum, and A. Sivaraman, 18th USENIX Symposium on Networked Systems Design and Implementation (NSDI 21).
• NSDI '19 "SIMON: A Simple and Scalable Method for Sensing, Inference and Measurement in Data Center Networks", Y. Geng, S. Liu, Z. Yin, A. Naik, B. Prabhakar, M. Rosenblum, and A. Vahdat, 16th USENIX Symposium on Networked Systems Design and Implementation (NSDI 19).
• NSDI '18 "Exploiting a natural network effect for scalable, fine-grained clock synchronization", Y. Geng, S. Liu, Z. Yin, A. Naik, B. Prabhakar, M. Rosenblum, and A. Vahdat, 15th USENIX Symposium on Networked Systems Design and Implementation (NSDI 18).
• Allerton "Self-programming networks: Architecture and algorithms", Y. Geng, S. Liu, F. Wang, Z. Yin, B. Prabhakar and M. Rosenblum, 2017 55th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

For a period in the past, I also worked on microwave-based medical imaging systems.

• "Beamforming Microwave-Induced Thermoacoustic Imaging for Screening Applications", H. Nan, S. Liu, J. G. Buckmaster, and A. Arbabian, IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 1, pp. 464-474, Jan. 2019.
• "Extracting Dielectric Spectroscopic Properties from Microwave-Induced Thermoacoustic Signals", S. Liu, H. Nan, N. Dolatsha, and A. Arbabian, The 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC’16), Orlando, Aug. 2016.
• "A 16-Element Wideband Microwave Applicator for Breast Cancer Detection Using Thermoacoustic Imaging", H. Nan, S. Liu, N. Dolatsha, and A. Arbabian, Progress in Electromagnetics Research Symposium (PIERS) Proceedings, 243-247, July 6-9, Prague, 2015. (Best Student Paper Award)

• Doctor of Philosophy (PhD), Department of Electrical Engineering, Stanford University, USA. 2015 - 2021
• Master of Science (M.S.), Department of Electrical Engineering, Stanford University, USA. 2015 - 2018
• Bachelor of Engineering (B.Eng.), Department of Electronic Engineering, Tsinghua University, Beijing, China. 2011 - 2015

• Staff Research Engineer, Clockwork Systems, Palo Alto, CA, 07/2021 -- Now
  Clockwork Systems (formerly known as Tick Tock Networks) is the startup commercializing our works of clock sync, network telemetry, congestion management, and more.
• Software Engineer, NetInfra Group, Google, Sunnyvale, CA, 04/2021 -- 07/2021
• Software Engineering Intern, Clockwork Systems, Palo Alto, CA, 06/2019 -- 09/2019
• Research Intern, Platform Group, eBay, San Jose, CA, 07/2018 -- 09/2018
• Software Engineering Intern, NetInfra Group, Google, Sunnyvale, CA, 07/2017 -- 09/2017
• Software Engineering Intern, Amazon Lab126, Sunnyvale, CA, 06/2016 -- 09/2016

• New York Times, on front page of 06/30/2018, "Whittling a New York Minute To 100 Billionths of a Second"
• New York Times, "Time Split to the Nanosecond Is Precisely What Wall Street Wants"
• Stanford News on July 19, 2018
• Facebook Research Blog, "Q&A with Stanford Professor Balaji Prabhakar, longtime university collaborator with Facebook" (covering the On-Ramp paper at NSDI 21 and the evaluation of On-Ramp in Facebook network infrastructure).
• TechCrunch, "Clockwork's Cloud Deluxe platform eliminates packet drops, improves cloud network performance".
• PR Newswire, "Clockwork Eliminates Packet Drops and Improves Cloud Network Performance with Launch of Cloud Deluxe Platform".