Collaborative Research: SI2-SSI: Data-Intensive Analysis for High Energy Physics (DIANA/HEP)

University Of Nebraska-Lincoln, Lincoln NE

Investigators

Abstract

Advanced software plays a fundamental role for large scientific projects. The primary goal of DIANA/HEP (Data Intensive ANAlysis for High Energy Physics) is developing state-of-the-art tools for experiments that acquire, reduce, and analyze petabytes of data. Improving performance, interoperability, and collaborative tools through modifications and additions to packages broadly used by the community will allow users to more fully exploit the data being acquired at CERN's Large Hadron Collider (LHC) and other facilities. These experiments are addressing questions at the heart of physics: What are the underlying constituents of matter? And how do they interact? With the discovery of the Higgs boson in 2012, the Standard Model of particle physics is complete. It provides an excellent description of known particles and forces. However, the most interesting questions remain open: What is the dark matter which pervades the universe? Does space-time have additional symmetries or extend beyond the 3 spatial dimensions we know? What is the mechanism stabilizing the Higgs boson mass from enormous quantum corrections? The next generation of experiments will collect exabyte-scale data samples to provide answers. Analyzing this data will require new and better tools. First, the project will provide the CPU and IO performance needed to reduce the iteration time so crucial to explore new ideas. It will develop software to effectively exploit emerging many- and multi-core hardware. It will establish infrastructure for a higher-level of collaborative analysis, building on the successful patterns used for the Higgs boson discovery and enabling a deeper communication between the theoretical community and the experimental community. DIANA?s products will sit in the ROOT framework, already used by the HEP community of more than 10000 particle and nuclear physicists. By improving interoperability with the larger scientific software ecosystem, DIANA will incorporate best practices and algorithms from other disciplines into HEP. Similarly, the project will make its computing insights, tools, and novel ideas related to collaborative analysis, standards for data preservation, and best practices for treating software as a research product available to the larger scientific community. Finally, to improve the quality of the next generation of software engineers in HEP, DIANA will host an annual workshop on analysis tools and establish a fellowship program.