Jianru (Jerry) Ding ☕️
Jianru (Jerry) Ding

PhD Student

About Me

Fifth-year PhD at the University of Chicago advised by Prof. Hank Hoffmann. Received two B.S. in Computer Science and Engineering and Finance from The Ohio State University and fortunate to work with Prof. Christopher Stewart.

Research Interest: My research interest lies in the joint of HPC, Computer Architecture and Operating Systems, and Machine Learning. My work focuses on control systems that adapt computing resource management to large-scale workload fluctuations to meet high-level user-defined goals.

Download CV
Interests
  • High Performance Computation
  • Computer Systems
  • Machine Learning
  • Computer Architecure
Education
  • PhD in Computer Science

    Unversity of Chicago

  • B.S. in Computer Science and Engineering

    The Ohio State University

  • B.S. in Finance

    The Ohio State University

📚 Research Projects

The UpDown system project

The Updown system is a large-scale distributed computing system that enables flexible graph representation and programmable intelligence to move it within the system. We build and explore efficient and scalable computation over massive graph structures in this system.

My interest and work falls into the intelligent and adaptive data movement and assignment to solve large-scale load-balancing problems. (ongoing work)

Distributed power management (SC’ 23)

The increase in modern cluster scale following Moore’s law is leaving power support behind. Fully powering an exascale cluster is neither realistic nor safe. Thus overprovisioned systems, clusters under a power cap, are proposed for exascale clusters to operate functionally. We explored distributed power management for such large-scale clusters.

Cache allocation performance modeling (ICAC’ 19)

Recent CPU architecture allows for customized cache allocation strategies. This opens new opportunities for computational sprinting to meet high-level user-defined goals.

Recent Publications
(2024). Efficiently Exploiting Irregular Parallelism Using Keys at Scale.
(2024). UpDown: Programmable fine-grained Events for Scalable Performance on Irregular Applications. arXiv preprint arXiv:2407.20773.
(2023). DPS: Adaptive Power Management for Overprovisioned Systems. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis.
(2019). Characterizing service level objectives for cloud services: Realities and myths. 2019 IEEE International Conference on Autonomic Computing (ICAC).