High Performance Computing and Seismic Data Working Group

Charge to the IRIS HPCWG

As the volume of archived seismic data increases, the need to have these data processed in new and more powerful computational systems has become more important. The High Performance Computing and Seismic Data Working Group (HPCWG) is a working group reporting to the IRIS Data Services Standing Committee (DSSC) that will focus on the use of the “Big Data” available in the IRIS DMC storage systems within high performance computing environments. The HPCWG will address data-driven seismological research requiring HPC resources, either for data processing or for simulation-based data assimilation. 

HPCWG Tasks

  1. Identify categories of data-driven seismological problems needing HPC
  2. Identify data accessibility methods and the performance needed to support the processing of seismic data in an HPC environment
  3. Define specific scenarios (“use cases”) requiring the use of HPC environments on the data at the IRIS DMC
  4. Recommend formats for observed seismic data that are suitable for use in an HPC environment.
  5. Identify possible partnerships between IRIS and HPC environments within the NSF (Computer and Information Science and Engineering [CISE], Extreme Science and Engineering Discovery Environment [XSEDE]), Department of Energy (DoE), or other agencies that have resources that can be shared with the academic research community.
  6. Report activities of the working group in advance of each DSSC meeting.

Membership

  • Carl Tape, Chair and member of the DSSC
  • Heiner Igel, Ludwig Maximilian University of Munich (LMU)
  • Weisen Shen, Washington University
  • Jeroen Tromp, Princeton University
  • Felix Waldhauser, Columbia University
  • Omar Ghattas, University of Texas 

Ex-Officio:

  • Arthur Rodgers, Lawrence Livermore National Lab (LLNL)
  • Stan Ruppert, Lawrence Livermore National Lab (LLNL)
  • Carene Larmat, Los Alamos National Lab (LANL)
  • Scott Klaske (ORNL)
  • Lorraine Hwang, Computational Infrastructure for Geodynamics
  • Tim Ahern, IRIS 

Term: 01 April 2016 to 30 July 2017

All meetings will take place virtually, using phone or internet communication.

Links and Downloads

Summary report of "Survey of Computational Needs for Data and Modeling in Seismology"

Relevant 2014 report, "Advancing solid earth system science through high-performance computing"