Engage second wave ECP/CEED applications

Technical Report · Fri Dec 21 04:00:00 EST 2018

DOI:https://doi.org/10.2172/1845634· OSTI ID:1845634

Min, Misun ^[1]; Camier, Jean-Sylvain ^[1]; Fischer, Paul ^[1]; Karakus, Ali ^[1]; Kerkemeir, Stefan ^[1]; Kolev, Tzanio ^[1]; Lan, Yu-Hsiang ^[1]; Medina, David ^[1]; Merzari, Elia ^[1]; Obabko, Aleks ^[1]; Ratnayaka, Thilina ^[1]; Shaver, Dillon ^[1]; Tomboulides, Ananias ^[1]; Tomov, Vladimir ^[1]; Warburton, Tim ^[1]

---

1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

The CEED co-design center aims to impact a wide range of ECP application teams through focused 1-to-1 and 1-to many interactions, facilitated by CEED application liaisons, ultimately targeted to provide easy-to-use discretization libraries for high-order FEM-based application simulations. The goal of this milestone was to extend our reach by identifying a second wave of ECP applications that can benefit from the efficient discretization algorithms and software developed in CEED. These applications were added to the list of the CEED first wave ECP applications (ExaSMR and MARBL), which continue to be main targets for our work. As part of the milestone, we also appointed liaisons on the CEED team that will be responsible for engaging each application, identifying its challenging discretization needs, and outlining a plan for interaction with CEED. Working with all CEED thrusts, the liaisons will be ultimately responsible for ensuring the adoption of CEED technologies in each application as well as for demonstrating their benefits. The selected second wave applications and CEED liaisons are: Urban - liaison Aleks Obabko (ANL), ExaWind - liaison Paul Fischer (UIUC), and ExaAM - liaison Tzanio Kolev (LLNL). In addition to details and results from these efforts, in this document we report on other project-wide activities performed in Q1 of FY19, including: engagements with performing scaling on V100 GPUs, collaboration with ECP/MPICH, participation in ECP/ST’s E4S release, the Laghos-2.0 release, papers, presentations, and other outreach efforts.

• Research Organization: Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
• DOE Contract Number: AC52-07NA27344
• OSTI ID: 1845634
• Report Number(s): LLNL-TR-764597; 954646
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING