Ke Fan
P.h.D candidate, Computer Science
University of Illinois Chicago
About me
As a Ph.D. candidate in Computer Science at the University of Illinois Chicago, I am conducting research under my advisor, Sidharth Kumar. My research lies in high-performance computing (HPC), emphasizing three key areas: optimizing the performance of MPI collectives, enhancing the performance of irregular parallel I/O operations, and improving the scalability of performance introspection frameworks.
Throughout my Ph.D. journey, I have made significant contributions to the field of high-performance computing, as evidenced by my publications in top-tier HPC conferences such as HPDC, HiPC, and ISC. My poster presentation at SC 2023 was recognized as the best poster finalist. I am also exploring new research areas, including GPU-aware collectives, auto performance tuning for collectives using ML, and ensemble performance visualization tools.
I was recently awarded the 2024 ACM/IEEE-CS George Michael Memorial High Performance Computing Fellowship.
Selected Publications
Configurable Algorithms for All-to-all Collectives
Ke Fan, Steve Petruzza, Thomas Gilray, Sidharth Kumar
The International Supercomputing Conference (ISC 2024)
Acceptance rate: 29%
Visit the paper for more information.
TinyProf: Towards Continuous Performance Introspection through Scalable Parallel I/O.
Ke Fan, Suraj Kesavan, Steve Petruzza, Sidharth Kumar
The International Supercomputing Conference (ISC 2024)
Acceptance rate: 29%
Visit the paper for more information.
Optimizing the Bruck Algorithm for Non-uniform All-to-all Communication.
Ke Fan, Thomas Gilray, Valerio Pascucci, Xuan Huang, Kristopher Micinski, Sidharth Kumar
International Symposium on High-Performance Parallel and Distributed Computing (HPDC 2022)
Acceptance rate: 19%
Visit the paper for more information.
Load-balancing Parallel I/O of Compressed Hierarchical Layouts.
Ke Fan, Duong Hoang, Steve Petruzza, Thomas Gilray, Valerio Pascucci, Sidharth Kumar.
IEEE Conference On High Performance Computing, Data, and Analytics (HiPC 2021)
Acceptance rate: 23%
Visit the paper for more information.
Exploring MPI Collective I/O and File-per-process I/O for Checkpointing a Logical Inference task.
Ke Fan, Kristopher Micinski, Thomas Gilray, Sidharth Kumar.
IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW 2021)
Visit the paper for more information.