Research Scientist
(formerly known as Facebook, Inc.)
E-mail:
|
|
Biography
Ke (David) is a research scientist in the Fault Tolerance Team of Meta (formerly known as Facebook). His current work focuses on building reliable and resilient software foundation for Meta’s private cloud on top of Meta’s geo-distributed data center infrastructure (across millions of machines in tens of geo-replicated data center regions) to serve billions of users globally and meet company's business needs.
He earned his PhD in Computer Science and Engineering (advisor Prof. Z. Morley Mao) from University of Michigan.
His thesis focuses on validating reliability, security and performance properties of mobile systems and emerging autonomous vehicle platforms using systematic software engineering techniques, including runtime profiling, program analysis and formal methods.
He received an MPhil (Master of Philosophy) and bachelor degree (first-class honour) in Computer Science and Engineering from The Hong Kong University of Science and Technology (HKUST).
His prior research also includes performance diagnosis and acceleration of Android systems, traffic engineering in software-defined networks, network and compute acceleration of cloud data centers.
Education
- University of Michigan, Ann Arbor
Ph.D. in Computer Science and Engineering
- The Hong Kong University of Science and Technology, Hong Kong
MPhil in Computer Science and Engineering
- The Hong Kong University of Science and Technology, Hong Kong
BEng in Computer Science, Computer Engineering (first class honours)
Research Work
- Performance, fault tolerance and security of cloud data center systems
1) Overload protection for large-scale microservices running in geo-distributed systems infrastructure [Defcon (OSDI 2023 paper)]
2) Disaster recovery for geo-distributed data center infrastructure. [Project Storm reported by Forbes][Storm drills experience to help recovery from Meta's global outage in October 4, 2021]
3) A datacenter-scale distributed virtual memory system that unifies the physical memory and disk resources on many compute nodes to form a system-wide data substrate for supporting large-scale in-memory computation. [VOLUME (CloudCom'13 paper)][Commercialization]
4) Cluster-area sensor network for cluster management and security enhancement in data centers. [MASS 2013 paper]
5) Wirelessly Assisted TCP for High-Bandwidth Data Center Networks. [MPhil Thesis]
- Safeguarding autonomous vehicle (AV) software platforms [2019 MTRAC Transportation award for AVGuardian][NSF SBIR awarded spin-off TrustedAVWare]
1) Developed automated program analysis and instrumentation infrastructure to verify and enforce safety policy compliance in AV software (e.g., Baidu Apollor, Autoware). [AVChecker (SIGMETRICS 2021 paper)][Traffic rule violation demo], [AVMaestro (IV 2022 paper)]
2) Developed customized program analysis and instrumentation infrastructure for AV software platforms (e.g., Baidu Apollo, Autoware) to discover security vulnerability and enforce policy-based permission control to mitigate cyberattacks against AV systems. [AVGuardian (EuroS&P 2020 paper)][Attack demo], [AVMaestro (IV 2022 paper)]
- Performance of mobile computing systems
1) Lightweight trace collection on mobile platforms and systematic, automated machine learning based characterization approach to facilitate root cause diagnosis of performance issues with mobile apps. [Internet-QoE 2016 paper][Media coverage], [PerfProbe (MOBILESoft 2019 paper)][Real-world Application]
2) Accelerating mobile applications through flip-flop replication. [Tango (MobiSys 2015 paper)][Software]
- Security enhancement by protocol customization: Program analysis and transformation to localize and disable optional vulnerable features in network and system protocol implementation for attack surface reduction. [FEAST 2017 paper][DoD STTR awarded technology transfer]
- Hybrid SDN: Incremental deployment of SDN in enterprise and ISP network systems for centralized traffic engineering. [SOSR 2016 paper][Pending Patent]
Publications
Peer-reviewed conference & workshop papers
-
Justin J. Meza, Thote Gowda, Ahmed Eid, Tomiwa Ijaware, Dmitry Chernyshev, Yi Yu, Md Nazim Uddin, Rohan Das, Chad Nachiappan, Sari Tran, Shuyang Shi, Tina Luo, David Ke Hong, Sankaralingam Panneerselvam, Hans Ragas, Svetlin Manavski, Weidong Wang, and Francois Richard. Defcon: Preventing Overload with Graceful Feature Degradation.
In Proceedings of the 17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2023). Jul. 10-12, 2023. Boston, MA.
-
Ze Zhang, Sanjay Sri Vallabh Singapuram, Qingzhao Zhang, David Ke Hong, Brandon Nguyen, Z. Morley Mao, Scott Mahlke, Qi Alfred Chen. AVMaestro: A Centralized Policy Enforcement Framework for Safe Autonomous-driving Environments.
In Proceedings of the 2022 IEEE Intelligent Vehicles Symposium (IV 2022). Jun. 5-9, 2022. Aachen, Germany.
-
Qingzhao Zhang, David Ke Hong, Ze Zhang, Qi Alfred Chen, Scott Mahlke, Z. Morley Mao. A Systematic Framework to Identify Violations of Scenario-dependent Driving Rules in Autonomous Vehicle Software.
In Abstract Proceedings of the 47th International Conference on Measurement and Modeling of Computer Systems (ACM SIGMETRICS 2021). Jun. 14-18, 2021. Beijing, China.
Video demo of simulated collision to pedestrian at crosswalk due to traffic rule violation, spin-off technology received NSF SBIR Phase I award.
-
David Ke Hong, John Kloosterman, Yuqi Jin, Yulong Cao, Qi Alfred Chen, Scott Mahlke, Z. Morley Mao. AVGuardian: Detecting and Mitigating Publish-Subscribe Overprivilege for Autonomous Vehicle Systems.
In Proceedings of the 5th IEEE European Symposium on Security and Privacy (EuroS&P 2020). Sep. 7-11, 2020. Genova, Italy.
Video demo of simulated attacks exploiting discovered vulnerability, filed an invention disclosure, received 2019 Michigan Translational Research and Commercialization award.
-
David Ke Hong, Ashkan Nikravesh, Z. Morley Mao, Mahesh Ketkar, Michael Kishinevsky. PerfProbe: A Systematic, Cross-Layer Performance Diagnosis Framework for Mobile Platforms.
In Proceedings of the 2019 IEEE/ACM 6th International Conference on Mobile Software Engineering and Systems (MOBILESoft 2019). May 25-26, 2019. Montréal, Canada.
Acknowledgement for PerfProbe's diagnosis findings from iNaturalist app developer team.
-
David Ke Hong, Qi Alfred Chen, Z. Morley Mao. An Initial Investigation of Protocol Customization.
In Proceedings of the ACM CCS Workshop on Forming an Ecosystem Around Software Transformation (FEAST 2017). Nov. 3, 2017. Dallas, TX.
Proposed technologies being commercialized through Navy-funded STTR project collaboration with P&J Robinson Corp.
-
Ashkan Nikravesh, David Ke Hong, Qi Alfred Chen, Harsha V. Madhyastha, Z. Morley Mao. QoE Inference Without Application Control.
In Proceedings of the ACM SIGCOMM Workshop on QoE-based Analysis and Management of Data Communication Networks (Internet-QoE). Aug. 26, 2016. Florianópolis, Brazil.
Media coverage in The Conversation.
-
David Ke Hong, Yadi Ma, Sujata Banerjee, Z. Morley Mao. Incremental Deployment of SDN in Hybrid Enterprise and ISP Networks.
In Proceedings of the 2nd ACM Symposium on SDN Research (SOSR 2016). Mar. 14-15, 2016. Santa Clara, CA.
Pending Patent, discussed in multiple popular surveys on hybrid SDN.
-
Mark Gordon, David Ke Hong, Peter Chen, Jason Flinn, Scott Mahlke, Z. Morley Mao. Accelerating Mobile Applications through Flip-Flop Replication.
In Proceedings of the 13th International Conference on Mobile Systems, Applications, and Services (MobiSys 2015). May 18-22, 2015. Florence, Italy.
ACM MobiSys Best paper award (2 out of 219 submissions), invited for research highlights in GetMobile magazine, included in the lecture material of a graduate course at Columbia University.
-
Zhiqiang Ma, Ke Hong, Lin Gu. VOLUME: Enable Large-Scale In-Memory Computation on Commodity Clusters.
In Proceedings of the 5th IEEE International Conference on Cloud Computing Technology and Science (CloudCom 2013). Dec. 2-5. 2013. Bristol, UK. [Errata]
Proposed technologies being commercialized in Data Thinker System by Hututa Technologies Limited.
-
Ke Hong, Shuo Yang, Zhiqiang Ma, Lin Gu. A Synergy of the Wireless Sensor Network and the Data Center System.
In Proceedings of the 10th IEEE International Conference on Mobile Ad-hoc and Sensor Systems (MASS 2013). Oct. 14-16, 2013. Hangzhou, China.
Journal papers
-
Qingzhao Zhang, David Ke Hong, Ze Zhang, Qi Alfred Chen, Scott Mahlke, Z. Morley Mao. A Systematic Framework to Identify Violations of Scenario-dependent Driving Rules in Autonomous Vehicle Software.
In Proceedings of the ACM on Measurement and Analysis of Computing Systems, Volume 5, Issue 2, Article 15, June 2021.
-
Mark Gordon, David Ke Hong, Peter Chen, Jason Flinn, Scott Mahlke, Z. Morley Mao. Tango: Accelerating Mobile Applications through Flip-Flop Replication.
In GetMobile: Mobile Computing and Communications, Volume 19, Issue 3, July 2015.
Poster
Thesis
Awards, Grants, Scholarship
Academic Services
- Technical Program Committee Member for IEEE/ACM International Conference on Mobile Software Engineering and Systems (MOBILESoft 2023, 2022 Technical Papers track, 2022 Industry Forum track)
- Artifact Evaluation/Review Committee Member for Privacy Enhancing Technologies Symposium (PETS 2023, 2022), 17th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2023), 2023 USENIX Annual Technical Conference (USENIX ATC '23), 17th European Conference on Computer Systems (EuroSys 2022), 28th ACM Symposium on Operating Systems Principles (SOSP 2021)
- Conference Paper Reviewer for Privacy Enhancing Technologies Symposium (PETS 2023, 2022), 33rd IEEE Intelligent Vehicles Symposium (IV 2022), 2021 International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME), 2022/2021 International Conference on Electrical, Computer and Energy Technologies (ICECET)
- Journal Reviewer for Journal of Industrial Information Integration (IF 10.063), IEEE Journal on Selected Areas in Communications (IF 9.144), IEEE Journal of Biomedical and Health Informatics (IF 5.772), IEEE Transactions on Mobile Computing (IF 5.577), IEEE Communication Letters (IF 3.436), Concurrency and Computation: Practice and Experience (IF 1.536).
- Student Volunteer for ACM SIGCOMM 2013, D-CLOUD 2011
Teaching
Ke worked as a teaching assistant or instructor for following undergraduate/graduate-level computer science courses at University of Michigan and Hong Kong University of Science Technology.
Talks and Demos
-
AVGuardian: Detecting and Mitigating Publish-Subscribe Overprivilege for Autonomous Vehicle Systems. Presentation at the 5th IEEE European Symposium on Security and Privacy (EuroS&P 2020). Sep. 10, 2020. Virtual event. [Talk]
-
PerfProbe: A Systematic, Cross-Layer Performance Diagnosis Framework for Mobile Platforms. Presentation at the 2019 IEEE/ACM 6th International Conference on Mobile Software Engineering and Systems (MOBILESoft 2019). May 25, 2019. Montréal, Canada. [Slides]
-
Performance, Security, and Safety Requirements Testing for Smart Systems Through Systematic Software Analysis. PhD thesis defense at University of Michigan. May 15, 2019. Ann Arbor, MI. [Slides]
-
An Initial Investigation of Protocol Customization. Presentation at the ACM CCS Workshop on Forming an Ecosystem Around Software Transformation (FEAST 2017). Nov. 3, 2017. Dallas, TX. [Slides]
-
A Synergy of Wireless Sensor Networks and Data Center Systems. MPhil thesis defense at the Hong Kong University of Science and Technology. Dec. 17, 2013. Hong Kong, China. [Abstract] [Slides]
-
A Synergy of the Wireless Sensor Network and the Data Center System. Presentation at the 10th IEEE International Conference on Mobile Ad-hoc and Sensor Systems (MASS 2013). Oct. 14-16, 2013. Hangzhou, China. [Slides]
-
Layer Zero: An Efficient Platform for Cloud Computing. Demo at China Hi-Tech Fair 2012 (CHTF 2012). Nov. 16-21, 2012. Shenzhen, China. [Poster (in Chinese)]
Work Experience
- Research Fellow, EECS Department, University of Michigan, Ann Arbor, MI 06/2019 - 06/2020
- Researched on program analysis and formal verification to validate security & safety of autonomous vehicle software
- Led commercialization of the developed technologies supported by MTRAC award
- Published research papers to IEEE EuroS&P 2020, ACM SIGMETRICS 2021 and IEEE IV 2022
- Strategic CAD Labs Research Intern, Intel Corporation, Folsom, CA 05/2016 - 09/2016
- Developed an app and OS-layer runtime profiling and statistical characterization prototype to diagnose performance problems with Android apps and applied it to guide the performance problem analysis and fixing in popular Android apps
- Published the work at the 6th IEEE/ACM International Conference on Mobile Software Engineering and Systems (MOBILESoft 2019)
- HP Labs Networking Research Associate, Hewlett Packard Enterprise, Palo Alto, CA, 06/2015 - 10/2015
- Designed a hybrid software-defined networking architecture for enterprise and ISP network systems
- Published the work at the 2nd ACM Symposium on SDN Research (SOSR 2016)
- Research Assistant, Department of Computer Science and Engineering, HKUST, 02/2013 - 05/2013
- Designed and implemented a secondary index for an exascale key-value storage system KV+
- Developed tutorials for Layer Zero and a web interface for KV+
- Teaching Assistant, Department of Computer Science and Engineering, HKUST, 09/2011 - 05/2013
- Conducted tutorials and labs of various courses for CSE majors
- Instructed linear algebra for freshmen in CSE department
- Junior Research Assistant, Department of Computer Science and Engineering, HKUST, 07/2011 - 08/2011
- Administrated a Xen-based research cluster
- Developed software VRS – a Xen-based virtual machine reservation system
Software