Enhancements supporting IC usage of PEM libraries on next-gen platforms

Technical Report · Wed Jun 23 00:00:00 EDT 2021

DOI:https://doi.org/10.2172/1798430· OSTI ID:1798430

Richards, David F. ^[1]; Ryujin, Brian S. ^[1]; Barton, Nathan ^[1]; Bastea, Sorin ^[1]; Beauchamp, Brock ^[1]; Beck, Bret ^[1]; Beckingsale, David ^[1]; Blake, Robert ^[1]; Brantley, Patrick ^[1]; Brown, Peter ^[1]; Burmark, Jason ^[1]; Carson, Robert ^[1]; Chen, Evelyn ^[1]; Collette, Mike ^[1]; Dawson, Shawn ^[1]; Fried, Larry ^[1]; Gert, Godfree ^[1]; Grondalski, John ^[1]; Gyllenhaal, John ^[1]; Hall, Burl ^[1] ; Haque, Riyaz ^[1]; Isaac, Ben ^[1]; Katz, M ^[2]; Kunen, Adam ^[1]; Kuo, I-F Will ^[1]; Le, Hai ^[1]; Loffeld, John ^[1]; Mattoon, Caleb ^[1]; McFadden, Marty ^[1]; McKinley, Scott ^[1]; Meraz-Rodriguez, Manny ^[1]; Miller, Doug ^[1]; Minner, Paul ^[1]; Nimmakayala, Rao ^[1]; Noble, Chad ^[1]; O'Brien, Matt ^[1]; Osawe, Maxwell ^[1]; Patel, Mehul ^[1]; Pozulp, Mike ^[1]; Pudliner, Brian ^[1]; Rana, Verinder ^[1]; Rieben, Rob ^[1]; Robinson, Peter ^[1]; Skinner, Aaron ^[1]; Slone, Dale ^[1]; Stephens, Branson ^[1]; Sterne, Philip ^[1]; Stevens, David ^[1]; Stitt, Tom ^[1]; Vargas, Arturo ^[1]; Wayne, Brett ^[1]; Weiss, Kenneth ^[1]; White, Chris ^[1]; Whitesides, Russell ^[1]; Yang, Max ^[1]; Yee, Ben ^[1]

---

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

This milestone reports on the culmination of several years of effort by multiple PEM support software development teams to provide capabilities for use in LLNL-developed integrated codes on next-gen ASC platforms, including GPU support. We will provide a survey of relevant Application Program Interfaces (API) that are required to support LLNL IC code capability on relevant architectures, with a focus on Sierra and El Capitan. We will identify and summarize all dependencies between PEM supported libraries and IC supported physics codes. We will provide an assessment of algorithmic improvements that have been deployed, as well as future developments that are required to complete the GPU porting efforts. This assessment will include a description of programming models adopted by each of the PEM projects, distinct algorithmic challenges for each of the capabilities, and information about sharing GPU memory between the APIs and host codes. We will develop targeted test problems to assess computational performance. Finally, this milestone will result in identification of gaps in our effort to assist the LLNL ASC program in prioritization of effort for porting software to El Capitan.

• Research Organization: Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
• Sponsoring Organization: USDOE National Nuclear Security Administration (NNSA)
• DOE Contract Number: AC52-07NA27344
• OSTI ID: 1798430
• Report Number(s): LLNL-TR-823775; 1036684
• Country of Publication: United States
• Language: English

Related Subjects

97 MATHEMATICS AND COMPUTING