TY - CONF AU - Kesselheim, Stefan AU - Herten, Andreas AU - Krajsek, Kai AU - Ebert, Jan AU - Jitsev, Jenia AU - Cherti, Mehdi AU - Langguth, Michael AU - Gong, Bing AU - Stadtler, Scarlet AU - Mozaffari, Amirpasha AU - Cavallaro, Gabriele AU - Sedona, Rocco AU - Schug, Alexander AU - Strube, Alexandre AU - Kamath, Roshni AU - Schultz, Martin G. AU - Riedel, Morris AU - Lippert, Thomas TI - JUWELS Booster – A Supercomputer for Large-Scale AI Research VL - 12761 M1 - FZJ-2021-04644 SN - 978-3-030-90538-5 (print) T2 - Lecture Notes in Computer Science SP - 453 - 468 PY - 2021 AB - In this article, we present JUWELS Booster, a recently commissioned high-performance computing system at the Jülich Supercomputing Center. With its system architecture, most importantly its large number of powerful Graphics Processing Units (GPUs) and its fast interconnect via InfiniBand, it is an ideal machine for large-scale Artificial Intelligence (AI) research and applications. We detail its system architecture, parallel, distributed model training, and benchmarks indicating its outstanding performance. We exemplify its potential for research application by presenting large-scale AI research highlights from various scientific fields that require such a facility. T2 - ISC High Performance 2021 CY - 24 Jun 2021 - 2 Jul 2021, Digital (Germany) Y2 - 24 Jun 2021 - 2 Jul 2021 M2 - Digital, Germany LB - PUB:(DE-HGF)8 ; PUB:(DE-HGF)7 UR - <Go to ISI:>//WOS:000763168300034 DO - DOI:10.1007/978-3-030-90539-2_31 UR - https://juser.fz-juelich.de/record/902889 ER -