The 1st internationaL workshop on Near Real-time Data Processing for Interconnected Scientific Instruments
September 16-17, 2024
Senri Life Science Center, Osaka Senri, Japan
In conjunction with IEEE eScience 2024
The complexity of scientific research calls for dynamic integration of various interconnected scientific instruments for data generation (e.g., experiments, observation, and simulation) and data analysis (e.g., AI/ML, visualization, etc.). The capability of near real-time data processing across interconnected scientific instruments is the foundation of various scientific workflows, including traditional human-in-the-loop and autonomous workflows. This is because analysis results are needed near real-time to provide time-sensitive decision-making and steering of experiments. However, as the improvement of scientific instruments leads to the generation of scientific data with unprecedented volumes and modalities, it imposes a huge strain on data processing as data acquisition, sharing, and analysis will be prohibitively expensive with the increase in data volumes. This landscape highlights the growing need for research efforts that optimize all stages of data processing at an extreme scale to enable near real-time processing, including but not limited to acquisition, reduction, management, storage, sharing, and analysis.
There are at least three important topics that our community is striving to answer: (1) how to design efficient data acquisition and reduction pipelines that support near-instrument preprocessing while maintaining the important features for scientific pursuit; (2) how to achieve extreme-scale data curation and sharing that leverages the advanced HPC infrastructure; (3) how to accommodate near real-time data analytics at extreme-scale with streaming or urgency requirements for time-sensitive decision making. Tackling these challenges requires expertise from computer science, mathematics, and application domains to study the problem holistically and develop solutions.
This international workshop targets HPC applications, researchers, and domain experts with big data problems and looking for new data management and analytical workflows for their applications. The outcome of this workshop will foster the implementation of near real-time data processing workflows by accelerating all stages of the scientific research lifecycle, including large-scale data acquisition, data curation, analytics, and sharing.
Workshop agenda (tentative)
Keynote speakers
Manish Parashar
University of Utah
Choong-Seock Chang
Princeton Plasma Physics Laboratory
Scott Klasky
Oak Ridge National Laboratory
Susumu Date
Osaka University
List of Topics
Data reduction methods for scientific data
Methods for diverse data types
Methods with accuracy- and feature-preserving guarantees
Optimal design of data reduction methods
AI4Compression
Data analysis over extreme-scale datasets
Surrogate/reduced-order models
Visualization techniques for near real-time computing
Metrics for reduction quality
Computing on reduced data
Accuracy and performance trade-offs
Reduction and system co-design
Accelerating reduction on emerging hardware
Data management and storage for near real-time computing
Runtime systems for data reduction
State of the practice
Resilient near real-time Computing
Submission guidelines
All papers must be original and not simultaneously submitted to another journal or conference. NRDPISI-1 will accept full papers in IEEE format following eScience formatting rules (limited to 6 pages excluding references) and extended abstracts (2 pages, except references and appendix).
Submission should be made to https://easychair.org/conferences/?conf=escience2024
IMPORTANT DATES
Full Paper submission deadline: July 29, 2024 (AOE)
Author notification: August 12, 2024
Camera-ready final submission deadline: August 23, 2024 (AoE)
Remote presentation submission deadline: August 30, 2024 (AoE)
Program committee (tentative)
TBD
Organization
Jieyang Chen, University of Alabama at Birmingham
Jinzhen Wang, Brooklyn College of City University of New York
Qing Liu, New Jersey Institute of Technology
Scott Klasky, Oak Ridge National Laboratory