Xinyi Chen Ph.D. Student Department of Computer and Information Science University of Pennsylvania E-mail: cxinyic at seas.upenn.edu Github: cxinyic CV (December 2021)

About Me

I'm a 4th-year Ph.D. student at the University of Pennsylvania working with Prof. Vincent Liu. My research interests include both databases and distributed systems. Currently, I am focusing on using new hardwares to offload network communications.

I got my bachelor degree from Shanghai Jiao Tong University and was a member of Advanced Network Laboratory (ANL).

Publications

• Cowbird: Freeing CPUs to Compute by Offloading the Disaggregation of Memory
  Xinyi Chen, Liangcheng Yu, Vincent Liu, and Qizhen Zhang
  To appear in SIGCOMM 2023
  


• Optimizing Data-intensive Systems in Disaggregated Data Centers with TELEPORT
  Qizhen Zhang, Xinyi Chen, Sidharth Sankhe, Zhilei Zheng, Ke Zhong, Sebastian Angel, Ang Chen, Vincent Liu, Boon Thau Loo
  ACM International Conference on Management of Data, SIGMOD 2022
  


• Understanding the Effect of Data Center Resource Disaggregation on Production DBMSs
  Qizhen Zhang, Yifan Cai, Xinyi Chen, Sebastian Angel, Ang Chen, Vincent Liu, Boon Thau Loo
  International Conference on Very Large Data Bases, VLDB 2020
  


• Provenance for Probabilistic Logic Programs
  Shaobo Wang, Hui Lyu, Jiachi Zhang, Chenyuan Wu, Xinyi Chen, Wenchao Zhou, Boon Thau Loo, Susan B. Davidson, Chen Chen
  International Conference on Extending Database Technology, EDBT 2020 (Best paper Award)


• R2-Tree: An Efficient Indexing Scheme for Server-Centric Data Center Networks
  Yin Lin, Xinyi Chen, Xiaofeng Gao, Bin Yao, Guihai Chen
  International Conference on Database and Expert Systems Applications, DEXA 2018
  

Education

• Ph.D. student: Sept. 2019 - present
  University of Pennsylvania
  Computer and Information Science (CIS)
  
• Bachelor: Sept. 2015 - June 2019
  Shanghai Jiao Tong University
  Computer Science, School of Electronic Information and Electrical Engineering (CS)
  
• Summer research intern: Jul. 2018 - Oct. 2018
  University of Pennsylvania
  Computer and Information Science (CIS)
  Supervisior: Prof. Boon Thau Loo
  

Selected projects

• DBMS in Disaggregated Data Center (DDC)
  
  Resource disaggregation is the new trend in data centers where compute, memory and storage node are separated and be connected by high speed network. There has not been a thorough evalution on the performance effects of DDC with data intensive applications. We run two popular open-source DBMSs, which are MonetDB and PostgreSQL and compare it with single-machine Linux. We find that there are significant performance degradation in disaggregation data center when running DBMS. Based on the finding, we propose some optimization methods.
• Provenance for Probabilistic Logic Programs
  
  We propose a novel provenance model called P3 for analyzing probabilistic logic programming (PLP) programs. P3 enables different kinds of queries: traditional explanation queries, queries for finding the set of most important derivations within an approximate error, top-K most influential queries, and modification queries that enable us to modify tuple probabilities with fewest modifications to program or input data. We apply these queries into real-world scenarios and the evaluation results show that the system can maintain provenance and execute queries efficiently with low overhead.

Awards & Honors

• Zhiyuan Honors Scholarship in Shanghai Jiao Tong University (top 5% in SJTU)
• Tung Scholarship in Shanghai Jiao Tong University (top 7 in computer science department)