UCSB Computer Science Powering Next Generation of Data Management Tools for MMA.

June 27, 2023

This article was originally published by Andrew Helregel, with the University of Illinois Urbana-Champaign's National Center for Supercomputing Applications (NCSA).

Read the full article here.

----------------------------------------------------------------------------------------------------------------------------------------------------

A proposal by the National Center for Supercomputing Applications (NCSA) and the Scalable Cyberinfrastructure to support Multi-Messenger Astrophysics (SCIMMA) group was approved by the National Science Foundation to develop communications infrastructure to coordinate and connect multi-messenger astrophysics research across the globe.

As the name suggests, multi-messenger astrophysics (MMA) is a field of study where astronomers are bringing together different “messengers,” or ways of studying the universe, such as photons, gravitational waves and high-energy particles. By combining these different data points, MMA can probe questions in fundamental physics through astrophysics. For example, the first and, to date, only known explosion associated with the merger of two neutron stars – a kilonova – detected by the LIGO-VIRGO-Kagra (LVK) Collaboration provided astronomers insight into how the heaviest elements in the universe were synthesized.

A schematic of the network SCIMMA is building. The Hopskotch network connects major facilities such as LIGO-Virgo-Kagra and the Vera C. Rubin Observatory, as well as other components that scientists use like the ANTARES alert broker and the NASA GCN Network. This allows these previously siloed facilities to communicate. Hopskotch users can identify the counterpart and trigger follow-up using space- and ground-based facilities around the world, reporting their results through the user-friendly, web-based HERMES service to the rest of the community using the Transient Name Server or the gravitational wave Treasure Map. SCIMMA also provides secure services for major experiments, allowing them to use the network for private internal communication. All data across SCIMMA’s network is archived and available through a database that respects the same security policies.

The new communications tools provided by this effort from NCSA and SCIMMA will allow astronomers and physicists to communicate more rapidly, robustly, and in a way that combines messages that both a human and a machine can read. For instance, physicists can send announcements of approximately where in the sky two neutron stars merged together. These can immediately be read by software, which triggers robotic telescopes, such as those of the Las Cumbres Observatory, to start observing the region, even if astronomers are asleep. Astronomers can then report their findings in a way that will instantly be accessible to other telescopes and facilities. The infrastructure is cloud-based, robust to single-point failures and has low latency. This allows for more-rapid action by astronomers to quickly observe changing phenomena before they disappear.

“More kilonovae could also teach us about the nature of dense nuclear matter, the formation and structure of jets and gamma-ray bursts, and the nature of gravity in the few seconds before a black hole is created in the merger, all while helping to resolve open questions in cosmology,” said Gautham Narayan, deputy director for the Center for AstroPhysical Surveys at NCSA and principal investigator of the project. “Coordinating multiple facilities to enable these kinds of discoveries is the key reason for SCIMMA.”

NCSA will lead a team of institutions for the more-than $3.5 million, four-year project working to build standardization between different subfields, as well as enable rapid response and global coordination of limited resources. This will necessitate a massive leap forward in the communications infrastructure currently employed. The team of co-PIs for the project includes Andy Howell of the Las Cumbres Observatory (LCO), Rich Wolski of the University of California Santa Barbara (UCSB), Adam Brazier of Cornell University and Chad Hanna of Penn State University.

Even a relatively modest investment can produce huge gains in the efficiency of billion-dollar investments for experiments such as the LIGO-VIRGO-Kagra Collaboration, the IceCube Neutrino Experiment and the Vera C. Rubin Observatory. This will have transformative effects in automating and speeding up both communications and research in other areas of time-domain science, including planet discovery, supernovae, Active Galactic Nuclei, Tidal Disruption Events, stellar variability and more. But the vision behind SCIMMA is even grander. What we are doing is really laying the foundation for a unified ecosystem for future astrophysical surveys. By breaking down the barriers between these experiments and building a shared infrastructure where researchers can collate, process and analyze data from all of these experiments together, we are working to enable science that simply has not been possible.
Gautham Narayan, Deputy Director for the Center for AstroPhysical Surveys at NCSA and PI of the project

“At UCSB, we are excited to be contributing state-of-the-art cloud computing research and research artifacts to help develop the next generation of tools for the MMA community,” said Wolski. “Science, at this scale, must be a team sport that takes advantage of the best results from technology disciplines like computer science. The SCIMMA effort, funded by the Cyberinfrastructure for Sustained Scientific Innovation (CSSI) program, is forward-looking in this regard, and we are pleased to be bringing UCSB’s computer science strength in scalable distributed systems to this leading-edge effort.”