High Performance Computing For Science

NASA meatball logo

Category: Instructional

  • Multiple Jobs per Node

    Background The number of CPUs (aka “cores”) per node among Discover’s processor architectures has continued to increase over time, with current Milan processors having 128 cores. Skylake and Cascade Lake nodes offer 40 and 46 cores per node respectively. Many NCCS users have legitimate use cases for significantly lower core counts per job (particularly for…

  • File Transfer from User to the Discover Cluster

    You must streamline your connections for direct SSH first before you can transfer files to the NCCS. For file transfers to ADAPT, please use the Transfer Nodes. For Command-Line Users In standard bastion configurations, you can run the following: If rsync is available on your client, it might be a better alternative to scp because you can resume…

  • NCCS Transfer Nodes

    The NCCS has configured a number of transfer nodes to improve our users’ ability to move medium and large datasets in and out of our systems as well as between NCCS systems. Currently the nodes allow transfers to, from, and between ADAPT and CSS storage. Small file transfers can to be performed on any node…

  • File Transfer from NCCS to User Systems

    You must streamline your connections for direct SSH first before you can transfer files from the NCCS. For file transfers from ADAPT, please use the Transfer Nodes. Command-Line Users In standard bastion configurations, you can run the following from your terminal: The ” . ” will place the file in your current working directory; you may specify a…

  • File transfer

    NCCS Transfer Nodes The NCCS has configured a number of transfer nodes to improve our users’ ability to move medium and large datasets in and out of our systems as well as between NCCS systems. Currently the nodes allow transfers to, from, and between ADAPT and CSS storage. Small file transfers can to be performed…

  • Storage and File System on ADAPT

    All file systems on ADAPT, including home and nobackup, are shared and accessible across all VM’s. File System Type Linux Filesystem Windows File System Default Quota Backup Cycles Home Directory PanFS /home H:\home 10GB Daily Scratch PanFS /explore/nobackup/people H:\nobackup\people 10TB No Backups Home Home directories will be available on all your virtual machines and have…

  • File System on Discover Cluster

    The Discover cluster provides several different types of file systems: home, nobackup, and temporary/scratch. See the showquota documentation for information on how to monitor your storage usage. File System Type Variable on Discover cluster Default Quota Backup Cycles Home Directory IBM GPFS $HOME 1GB Daily Scratch IBM GPFS $NOBACKUP 5Tb/300k inodes No Backups Scratch local $LOCAL_TMPDIR node…

  • Discover GPU Partition

    GPU Availability Within The Discover Cluster Scalable Unit 16 (SCU16) makes GPU resources available within the NCCS Discover cluster’s gpu_a100 partition, which comprises 10 AMD nodes that each include: Note: These nodes will be fully shared, with individual nodes running jobs belonging to multiple users. Each user is limited to a maximum of one node…

  • SSH, PIV, and other Connectivity Issues

    If you are having issues logging in to an NCCS system, it could be due to one of the following reasons: Resolving PIV SSH Issues PIV SSH is only possible if you have a NASA provided PIV card and PIV card reader. Most PIV SSH issues can be resolved by checking your client configuration. Please…

  • Discover SBU Rates

    The NCCS uses a Standard Billing Unit (SBU) for allocating and tracking computing time usage. The new SBU tracking system started to appear in usage reports October 2018. View the full set of SBU Conversion Factors and an explanation of changes on the NASA High-End Computing Website. ARCHITECTURE Supermicro TwinPro Supermicro TwinPro Supermicro FatTwin SGI Rackable…