Research Interest

• In-Situ Techniques
• Performance Modelling
• Dynamic Resource Management

Publications and Talks

Publications 

• Ju, Yi, Amir Raoofy, Dai Yang, Erwin Laure, and Martin Schulz. "Exploiting reduced precision for GPU-based Time series mining." In 2022 Springer IEEE International Parallel and Distributed Processing Symposium (IPDPS), pp. 124-134. IEEE, 2022.
• Ju, Yi, Adalberto Perez, Stefano Markidis, Philipp Schlatter, and Erwin Laure. "Understanding the impact of synchronous, asynchronous, and hybrid in-situ techniques in computational fluid dynamics applications." In 2022 IEEE 18th International Conference on e-Science (e-Science), pp. 295-305. IEEE, 2022.
• Ju, Yi, Mingshuai Li, Adalberto Perez, Laura Bellentani, Niclas Jansson, Stefano Markidis, Philipp Schlatter, and Erwin Laure. "In-Situ Techniques on GPU-Accelerated Data-Intensive Applications." In 2023 IEEE 19th International Conference on e-Science (e-Science), pp. 1-10. IEEE, 2023.
• Ju, Yi, Dominik Huber, Adalberto Perez, Philipp Ulbl, Stefano Markidis, Philipp Schlatter, Martin Schulz, Martin Schreiber, and Erwin Laure. "Dynamic Resource Management for In-Situ Techniques Using MPI-Sessions." In European MPI Users' Group Meeting, pp. 105-120. Cham: Springer Nature Switzerland, 2024.
• A Performance Model of In-Situ Techniques (Accepted by Parallel and Distributed Processing Conference 2025)

Talks 

• Ju, Yi, "High-Performance Mining of Multidimensional Time Series with Multiple GPUs and at Reduced Precision." at NVIDIA GTC 2021
• Ju, Yi, "Exploring in-situ techniques and resource allocation on Quantum ESPRESSO." at EURO-PAR 2023.
• Ju, Yi, "In-Situ Techniques with ADIOS2." at ISC 2024.