High Energy Physics Requirements Review Now Available: The Data Deluge Shows no Sign of Cresting!

Lauren Rotman and Jason Zurawski

Across the physical sciences, new instruments and capabilities are continuing a relentless growth in data production and need for high speed networking and analysis resources.

ESnet stays on-top of these trends via the Network Requirements Review process, which for the past 15 years has been a remarkable and useful collaboration between the DOE Office of Advanced Supercomputing Research (ASCR), ESnet and science programs across the DOE Office of Science.

The latest Network Requirements Review for the Office of Science High Energy Physics program office (HEP) is now available — among many other findings, this review confirms that the exponential growth of scientific data generation will continue unabated as we proceed into what may well be a new golden age for high energy physics research. Some samples include:

⇾ The upcoming High Luminosity era for the Large Hadron Collider (beyond 2027, or Run-4) will require multi-Tbps network speeds to support globally dispersed “Tier 1” HPC resources. Scientists will use the LHC to uncover how the Higgs-Boson interacts and gives mass to other particles, and explore emerging evidence for particle behaviors not explained by current physics models. Each data-taking year, the experiments, ATLAS and CMS combined, are expected to accumulate roughly 1 EB of new data and it is estimated that complete data set sizes may routinely exceed 100 PB.

*Expected maximum luminosity and integrated luminosity for the LHC as a function of* *calendar year, data produced tracks with improved luminosity and resolution*

⇾ Scientists at the Deep Underground Neutrino Experiment (DUNE) in South Dakota and at Fermilab in Illinois, will use high speed data transfer to identify supernova events, as part of ongoing measurement of neutrino interactions. Supernovae measured by DUNE will generate over 200TB of compressed data per event, and Research and Educational Networks (REN) must be able to supply highly reliable, predictable data transfer capabilities to provide telescope targeting data to global arrays.

10kt DUNE Far-Detector SP module, showing the alternating 58 m long (into the page), 12 m high anode (A) and cathode (C) planes, as well as the field cage that surrounds the drift regions between the anode and cathode planes. The blank area on the left side was added to show the profile of a single anode plane assembly (APA). Person included for scale.

⇾ The Cosmic Microwave Background, Stage 4 (CMB-S4) experiment will require data management and transfer capabilities in some of the most demanding locations on earth. Operating two observational locations, and multiple telescopes with a combined total of 500,000 cryogenically-cooled superconducting detectors at the South Pole and in the Chilean Atacama Desert, CMB-S4 will provide an unprecedented picture back into the start of the Universe. Operating for seven years in these conditions, 22 TB (~8 TB at the South Pole and ~14 TB in Chile) of data will be generated daily, leading to an accrual of 3 PB annually, and as much as 100 TB over the full program lifecycle.

Two Cross-Dragone (CD) telescopes (one is pictured above) with six meter diameter input apertures will be deployed at the Chilean site to map roughly 70% of the sky every day to support the dark universe, matter-mapping, and time-varying mm-wave sky science goals. Image and caption courtesy of the CMB-S4 Project

Network Requirements Reviews analyze the current, near, and long-term needs of the HEP community, providing a network and data-centric understanding of the scientific process used by the researchers and scientists. These requirements reviews drive ESnet’s investments in new services and capabilities, and enable ESnet to build strong partnerships with Office of Science (SC) programs, PIs, and user facilities. More information on this ESnet requirements review process can be found here.

We would like to thank the 13 HEP projects, and all of the HEP & DOE Office of Science collaborators who generously gave of their time, expertise, and most importantly, their enthusiasm for the future of high energy physics, as part of creating this report.

We want to especially thank the entire Science Engagement team plus Kate Robinson, and Dale Carder with our Network Engineering group who all provided outstanding support and technical expertise.