The Case for Columnar Analysis (a Two-Part Series) [PowerPoint]

Technical Report · Thu Mar 21 00:00:00 EDT 2019

DOI:https://doi.org/10.2172/1570209· OSTI ID:1570209

Smith, Nick ^[1]; Gray, Lindsey ^[1]; Cremonisi, Matteo ^[1]; Jayatilaka, Bo ^[1]; Gutsche, Oliver ^[1]; Hall, Allison ^[1]; Pedro, Kevin ^[1]; Melo, Andrew ^[2]; Pivarski, Jim ^[3]; Galewsky, Ben ^[4]; Neubauer, Mark ^[5]

---

1. Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
2. Vanderbilt Univ., Nashville, TN (United States)
3. Princeton Univ., NJ (United States)
4. Univ. of Illinois at Urbana-Champaign, IL (United States). National Center for Supercomputing Applications (NCSA)
5. Univ. of Illinois at Urbana-Champaign, IL (United States)

This talk provides prologue terminology and technology. Part I covers analyzer experience including user experience, code samples, domain of applicability, and scalability. Part II discusses technical underpinnings including theoretical motivation, the Coffea framework, factorized data delivery, package ecosystem, performance, and future directions.

• Research Organization: Fermi National Accelerator Lab. (FNAL), Batavia, IL (United States)
• Sponsoring Organization: USDOE Office of Science (SC), High Energy Physics (HEP)
• Contributing Organization: Coffea Team; Institute for Research and Innovation in Software for High Energy Physics (IRIS-HEP)
• DOE Contract Number: AC02-07CH11359
• OSTI ID: 1570209
• Report Number(s): FERMILAB-SLIDES-19-007-T; oai:inspirehep.net:1757703; TRN: US2100578
• Resource Relation: Conference: 2019 Joint HSF/OSG/WLCG Workshop (HOW 2019), Newport News, VA (United States), 18-22 Mar 2019
• Country of Publication: United States
• Language: English

Related Subjects

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS
97 MATHEMATICS AND COMPUTING