Preparing NERSC users for Cori, a Cray XC40 system with Intel many integrated cores. (25th August 2017)

• Record Type: Journal Article
• Authors: He, Yun; Cook, Brandon; Deslippe, Jack; Friesen, Brian; Gerber, Richard; Hartman‐Baker, Rebecca; Koniges, Alice; Kurth, Thorsten; Leak, Stephen; Yang, Woo‐Sun; Zhao, Zhengji; Baron, Eddie; Hauschildt, Peter
• Other Names: Xhafa Fatos guestEditor.; Michael Scott guestEditor.; He Yun guestEditor.
• Is Part Of: Concurrency and computation. Volume 30:Number 1(2018)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2017-08-25
• Subjects: cross compilation -- Intel Xeon Phi -- KNL -- performance optimization -- process and thread affinity -- user support; Parallel processing (Electronic computers) -- Periodicals; Parallel computers -- Periodicals; 004.35
• DOI: 10.1002/cpe.4291
• Languages: English
• ISSNs: 1532-0626

Abstract

Summary: The newest NERSC supercomputer Cori is a Cray XC40 system consisting of 2, 388 Intel Xeon Haswell nodes and 9, 688 Intel Xeon‐Phi "Knights Landing" (KNL) nodes. Compared to the Xeon‐based clusters NERSC users are familiar with, optimal performance on Cori requires consideration of KNL mode settings; process, thread, and memory affinity; fine‐grain parallelization; vectorization; and use of the high‐bandwidth MCDRAM memory. This paper describes our efforts preparing NERSC users for KNL through the NERSC Exascale Science Application Program, Web documentation, and user training. We discuss how we configured the Cori system for usability and productivity, addressing programming concerns, batch system configurations, and default KNL cluster and memory modes. System usage data, job completion analysis, programming and running jobs issues, and a few successful user stories on KNL are presented.