Jingyuan Zhu
Ph.D student in CS @ UMich
Ann Arbor, Michigan
Email
Github
Google Scholar
Hello, my name is Jingyuan Zhu. I’m a Ph.D Candidate at Computer Science Engineering at University of Michigan, advised by Prof. Harsha V. Madhyastha. My research interest is in Network Systems. Specifically, my research focuses on studying & characterization of broken links on the web, and deriving solutions for dealing with broken links with the leverage of web archives. Up to now, I’ve looked into the reason for broken references on Wikipedia, helped design a new web archive that saves storage and preserves better fidelity, and implemented a system that automatically finds new URLs for the same pages whose original URLs no longer resolve.
Before studying as a Ph.D in Michigan, I received BSE from UM in CS in a dual degree with BSE from Shanghai Jiao Tong University in ECE
News
Aug, 2023 | Our paper “Reviving Dead Links on the Web with FABLE” got accepted at IMC’23! Excited to go to Montréal this fall! |
---|---|
Aug, 2023 | Interned @ Google this summer. Had a great time in Seattle with the Browser Loading team (now MoQaTi)! |
Jul, 2023 | Our work Sprinter got accepted at NSDI’24! |
Nov, 2022 | Attended HotNets22. Enjoy in Austin, Texas! |
Publications
- Sprinter: Speeding up High-Fidelity Crawling of the Modern WebIn 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 2024). NSDI24
- Reviving Dead Links on the Web with FABLEIn Proceedings of the 23nd ACM Internet Measurement Conference. IMC23
- Making links on your web pages last longer than youIn Proceedings of the 21st ACM Workshop on Hot Topics in Networks. HotNets22
- Characterizing "permanently dead" links on WikipediaIn Proceedings of the 22nd ACM Internet Measurement Conference. IMC22
- Jawa: Web Archival in the Era of {JavaScript}In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22). OSDI22
- CPU microarchitectural performance characterization of cloud video transcodingIn 2020 IEEE International Symposium on Workload Characterization (IISWC). IISWC20