Continued advances in computational power and high-speed networking is
enabling a new model of scientific experiment, experiment-in-the-loop
computing (EILC). In this model, high-end computing systems are closely
coupled to experimental and observational infrastructure, and the two
interact to drive a deeper understanding of physical phenomena. At the
same time, advances and widespread adoption of machine learning enables
new ways to register and use experimental data.
Several research and development challenges are posed by this
multifaceted paradigm, many of which are independent of the particular
scientific application domain. New algorithms to integrate simulation
outputs and experimental data sets must be developed. High performance
data management and transfer techniques must be developed to manage and
manipulate simulated and observed data sets. Workflows must be
constructed with high levels of usability and understandability to
enable scientific post-analysis and improvement of the computing solution.
The Workshop on Experiment-in-the-Loop Computing (XLOOP 2023) will be a
unique opportunity to promote this cross-cutting, interdisciplinary
topic area. We invite papers, presentations, and participants from
the physical and computer sciences, and encourage the sharing of
ideas from across these domains to find common solutions and
technologies to make rapid progress in EILC, so that many application
areas can easily adopt these methods.
Topics of interest include, but are not limited to:
Machine learning applications in simulation or experiment control
Case studies in EILC applications and solutions
Data transfer techniques and technologies
In situ analysis methods and tools relevant to experiment data
Simulation and experiment validation methods and tools
Workflow technologies to manage computation and experiment couplings
Advanced systems architecture for EILC applications
High-performance I/O methods and libraries
Data integration and assimilation algorithms and technologies
Performance evaluation in EILC applications and solutions
Cyberinfrastructure and “big science” planning and reporting
Portable solutions for reproducible, transferable experiments