CompF1: Experimental Algorithm Parallelization

• Functional areas
  • Parallelization of Detector reconstruction algorithms, physics object reconstruction/calibration algorithms
  • Utilization of CPU, accelerator hardware and what comes next in 5-10 years
  • Developing better algorithms in addition to parallelization
  • Portability solutions that support the same algorithm implementation on multiple hardware architectures
• Mandate
  • Describe experimental algorithms that are or will be used by the stakeholders
  • What are the processing resource needs to execute the experimental algorithms of the stakeholders
  • What is the technology evolution of processing resources?
    • Coordinate with theoretical calculations and simulations working group
  • How will stakeholders be able to design and write algorithms for these processing resources?
    • Coordinate with theoretical calculations and simulations working group
    • Coordinate with Machine Learning working group
  • How can coding standards and performance standards be used?
    • Coordinate with theoretical calculations and simulations working group, and Collaboration and Ecosystem working group
  • How are the solutions used by the community embedded derived from solutions from industry/other science domains

The general time schedule can be found here: https://snowmass21.org/computational/start#time_schedule

• Nov. 2020 - Dec. 2020
  • December 1: Skeleton of document defining initial set of topics and key questions available
  • December 16th, 15:00 CT: CompF1 meeting to discuss topics and outline of document.

Here is the list of submitted LOIs to this topical group. First index before “/” corresponds to the primary frontier used for the submission.

A draft of the CompF1 report can be found here. It contains an outline with the main topics as well as a preliminary executive summary that is aimed at summarizing CompF1s contribution to the final CompF report. Suggestions and contributions to the report are welcome.