Share
Jean Luca Bez
Jean Luca is a Carrer-Track Researcher in the Scientific Data Management Group at Lawrence Berkeley National Laboratory (LBNL), USA. He is passionate about High-Performance I/O, Parallel I/O, Education, and Competitive Programming. His research focuses on optimizing the I/O performance of scientific applications at the middleware level by exploring I/O Forwarding, I/O Scheduling, and Automatic Tuning and Reconfiguration using Machine Learning techniques.
» Visit Jean Luca's personal web page
Journal Articles
Jean Luca Bez, Houjun Tang, Scot Breitenfeld, Huihuo Zheng, Wei-Keng Liao, Kaiyuan Hou, Zanhua Huang, Suren Byna, "h5bench: Exploring HDF5 Access Patterns Performance in Pre-Exascale Platforms", Concurrency and Computation: Practice and Experience (CCPE), January 31, 2024,
Jean Luca Bez, Suren Byna, Shadi Ibrahim, "I/O Access Patterns in HPC Applications: A 360-Degree Survey", ACM Computing Surveys, September 15, 2023, 56, doi: 10.1145/3611007
André Ramos Carneiro, Jean Luca Bez, Carla Osthoff, Lucas Mello Schnorr, Philippe O.A. Navaux, "Uncovering I/O demands on HPC platforms: Peeking under the hood of Santos Dumont", Journal of Parallel and Distributed Computing, August 18, 2023, 182, doi: https://doi.org/10.1016/j.jpdc.2023.104744
Conference Papers
Jean Luca Bez, Suren Byna, "Exploring the Proactive Data Containers Runtime System in VAST - A Case Study", 9th International Parallel Data Systems Workshop (PDSW), 2024,
Hiniduma, K., Byna, S., Bez, J. L., Madduri, R., "AI Data Readiness Inspector (AIDRIN) for Quantitative Assessment of Data Readiness for AI", 36th International Conference on Scientific and Statistical Database Management (SSDBM 2024), 2024,
Egersdoerfer, C., Sareen, Arnav., Bez, J. L., Byna, S., Dai, D., "ION: Navigating HPC I/O Optimization Journey using Large Language Models", 16th ACM Workshop on Hot Topics in Storage and File Systems (HotStorage'24), 2024, doi: 10.1145/3655038.3665950
Hammad Ather, Jean Luca Bez, Yankun Xia, Suren Byna, "Drilling Down I/O Bottlenecks with Cross-layer I/O Profile Exploration", 38th IEEE International Parallel & Distributed Processing Symposium, San Francisco, CA, USA, May 27, 2024,
Neeraj Rajesh, Keith Bateman, Jean Luca Bez, Suren Byna, Anthony Kougkas, Xian-He Sun, "TunIO: An AI-powered Framework for Optimizing HPC I/O", 38th IEEE International Parallel & Distributed Processing Symposium, San Fransicso, CA, US, May 27, 2024,
Jakob Luettgau, Shane Snyder, Tyler Reddy, Nikolaus Awtrey, Kevin Harms, Jean Luca Bez, Rui Wang, Rob Latham, Philip Carns, "Enabling Agile Analysis of I/O Performance Data with PyDarshan", Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis, Denver, CO, USA, Association for Computing Machinery, November 12, 2023, 1380–1391, doi: 10.1145/3624062.3624207
Bin Dong, Jean Luca Bez, Suren Byna, "AIIO: Using Artificial Intelligence for Job-Level and Automatic I/O Performance Bottleneck Diagnosis.", In Proceedings of the 32nd International Symposium on High-Performance Parallel and Distributed Computing (HPDC ’23), June 16, 2023,
- Download File: IODiagnose-final.pdf (pdf: 1.9 MB)
Hammad Ather, Jean Luca Bez, Boyana Norris, Suren Byna, "Illuminating the I/O Optimization Path of Scientific Applications", High Performance Computing: 38th International Conference, ISC High Performance 2023, Hamburg, Germany, May 21–25, 2023, Proceedings, Hamburg, Germany, Springer-Verlag, May 21, 2023, 22–41, doi: https://doi.org/10.1007/978-3-031-32041-5_2
The existing parallel I/O stack is complex and difficult to tune due to the interdependencies among multiple factors that impact the performance of data movement between storage and compute systems. When performance is slower than expected, end-users, developers, and system administrators rely on I/O profiling and tracing information to pinpoint the root causes of inefficiencies. Despite having numerous tools that collect I/O metrics on production systems, it is not obvious where the I/O bottlenecks are (unless one is an I/O expert), their root causes, and what to do to solve them. Hence, there is a gap between the currently available metrics, the issues they represent, and the application of optimizations that would mitigate performance slowdowns. An I/O specialist often checks for common problems before diving into the specifics of each application and workload. Streamlining such analysis, investigation, and recommendations could close this gap without requiring a specialist to intervene in every case. In this paper, we propose a novel interactive, user-oriented visualization, and analysis framework, called Drishti. This framework helps users to pinpoint various root causes of I/O performance problems and to provide a set of actionable recommendations for improving performance based on the observed characteristics of an application. We evaluate the applicability and correctness of Drishti using four use cases from distinct science domains and demonstrate its value to end-users, developers, and system administrators when seeking to improve an application’s I/O performance.
