Scholar
Haochen Pan
Google Scholar ID: XUkFYGQAAAAJ
University of Chicago
Citations & Impact
Citations
185
H-index
7
i10-index
5
Publications
20
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- Published works on Octopus and Icicle, with the latest research available as a preprint.
Research Experience
- Developed Octopus, a Kafka-based hierarchical event fabric that enables high-performance exchange of control and metadata events across cloud and HPC facilities. Designed Icicle (to be released), which integrates Octopus, Apache Flink, and Globus Search (backed by Elasticsearch) to monitor metadata updates for Lustre and IBM Storage Scale filesystems, providing real-time visibility to system administrators and enabling comprehensive historical usage evaluations. Recently, leveraged the Model Context Protocol (MCP) to expose research capabilities in heterogeneous cyberinfrastructure discoverable, invokable, and composable by LLM-powered agents.
Education
- Ph.D. student in Computer Science at the University of Chicago, advised by Dr. Ian Foster and Dr. Kyle Chard.
Background
- Research Interests: Distributed Systems, Cloud Computing, High-Performance Computing (HPC); Focus: Improving resilience and efficiency in AI-guided scientific experiments and distributed, time-sensitive data analysis.
