frost user document heading  
               
NCAR
Last update: 11/20/2007

Frost user doc contents

System information overview

This document provides the basic information new users need before they can start using this computer.

If you are new to computing at NCAR, you can greatly increase your ability to work productively here by reading Overview of computing at NCAR.

Frost is shared by several projects:

  1. Joint project between the University of Colorado and NCAR.
  2. As of August 1, 2007, Frost is also a TeraGrid computing resource. Twenty-five percent (25%) of frost cycles are available to all investigators with NSF allocations who wish to compute under the TeraGrid project, sponsored by the National Science Foundation.
  3. Each section of this document provides information needed to use frost in either of two ways: (1) under an NCAR project, or (2) under a TeraGrid project.

    System hardware

    Frost is a single-rack, IBM Blue Gene/L system with 1024 compute nodes. There is one I/O node for every 32 compute nodes, for a total of 32 I/O nodes in the rack. Each compute node and I/O node is a dual-core chip, containing two 700MHz PowerPC-440 CPUs, 512MB of memory, and two floating-point units (FPUs) per core. Thus frost has a total of 2048 processors capable of sustaining a peak performance of 5.734 trillion floating-point operations per second (TFLOPs). By default, the compute nodes run in coprocessor mode (one processor handles computation and the other handles communication). In virtual-node mode, both processors on a node are available to run user code, but the processors share both the computation and the communication load.

    The operating system is SuSE Linux Enterprise Server 9. A total of 6 TB of disk space is available.

    Recommended use

    The Frost Blue Gene/L system is a highly scalable platform for developing, testing and running parallel MPI applications up to 2048 processors, and providing efficient computing for smaller job sizes.

    More information about frost's system architecture appears in Figure 1. Technical schematic of frost BlueGene/L supercomputer

    Software environment

    • Operating System: Linux 2.6
    • Compilers: blrts_xlc, blrts_xlC, blrts_xlf, blrts_xlf90, blrts_xlf95
    • Profilers: HPM, mp_profilers, Peekperf, Xprofiler, gprof, mpi_trace
    • Debugger: Totalview
    • Utilities: gmake, gcc, contact the CISL consultants for more information
    • Software libraries:
      •  Mass

      •  netCDF

      •  Parallel-netCDF

      •  BLAS/LAPACK

      •  ESSL

    • Batch subsystem: Cobalt
    • System information commands: cqstat for queued and running job status, cqsub for job submission, partlist for partition information

    System security and access method

    NCAR frost users

    Users accessing frost from inside the UCAR security perimeter must have a CRYPTOCard. If you do not have a CISL-issued CRYPTOCard, you can request one by submitting an Extraview work request or calling 303-497-1200. Instructions for navigating the UCAR security perimeter are provided in the document "Overview of computing at NCAR" and the relevant section is Accessing computing services.

    You will need Secure SHell (SSH) on the workstation you use to access CISL systems. Information about using SSH for computing at NCAR is described in the document "Overview of computing at NCAR" and the relevant section is SSH - Secure SHell.

    Use the "secure" commands (scp and sftp) on frost to transfer files.

    TeraGrid frost users

    TeraGrid users may access frost using ssh, gsissh, or single-sign-on using certificates. Please see the section on Interactive access for details.

    File systems

    The General Parallel File System (GPFS) is installed on frost. The scratch space (/ptmp) is GPFS. Home directories are NFS mounted.

    Scratch space

    The large amounts of data generated by executables should be written to temporary disk in individual /ptmp directories (/ptmp/$LOGNAME, where $LOGNAME is the user login name). You may generate subdirectories on /ptmp and expect that /ptmp data will exist for the duration of the user job or interactive login. Valuable data on /ptmp should be moved to the MSS or other storage before a batch job or user login session terminates.

    The /ptmp file system is not backed up, and files are subject to deletion if the system gets too full.

    Table 1. Filesystems on Frost

    Path Soft Quota Hard Quota Use Backups Use
    /home/[username] 4MB 16MB binaries ~Monthly The /home filesystem is NFS mounted from the Blue Gene/L service node, and is not to be used for job data.
    /ptmp/[username] 600GB 800GB Job data Never /ptmp is a GPFS filesystem which provides up to 600MB/s for job I/O and job data. The filesystem is 6TB, and the quotas are only to prevent abuse - please police your own use. Files are not scrubbed at present, but under high use the administrative staff will contact the peak users, and may potentially intervene.
    /mnt/gpfs-wan (see SDSC) (see SDSC) Copy files to/from this data store and local resources
    The GPFS-WAN filesystem from SDSC is available as /mnt/gfps-wan, but is only provided as a convenience for moving files between the sites. Currently, running jobs out of this filesystem is not supported. Note that we do not copy the entire set of TeraGrid DN's - if your home directory on this filesystem shows up as being owned by 'nobody', the local set of DN's for your user account does not match the set used for GPFS-WAN. Please open a ticket at help@teragrid.org to have us add the additional DN's for your account.

    Obtaining your account and login

    Requesting an NCAR account

    Regular accounts on Frost may be available to investigators who are sponsored by one of the Co-Investigators on the grant used to purchase Frost.

    To get an account on Frost, have your sponsor send a request to Henry Tufo, with the following information:

    1. First name
    2. Last name
    3. Email address
    4. Telephone number
    5. FAX number
    6. Institution name if it is not apparent from the email address
    7. If they need a NCAR cryptocard, and a FedEx address if so. It needs to be a complete mailing address - FedEx will not accept P.O. Box or campus drop box.

    This request will be reviewed and if approved will be forwarded on to the appropriate NCAR team for account creation. You will receive an email notifying you of your new account, and a document providing an overview of Frost will be attached. A few days later you should receive your cryptocard at the FedEx address provided in the request. At this point you will be able to log in and use the system.

    Requesting a TeraGrid account

    Information on requesting a TeraGrid account is given at:

      http://www.teragrid.org/userinfo/access/allocations.php

    Roaming accounts are not automatically enabled. In order to access the Frost system, please send email to help@teragrid.org requesting that your roaming account be enabled.

    User environment - shells, paths, dotfiles

    Setting up your NCAR environment

    New users are placed in the bash shell by default. The csh, ksh, and a few other less common shells (see /etc/shells) are also available.

    Basic shell startup files are provided for all users at the time of account creation. To see the default environment for your login, type:

    printenv

    when you are logged in.

    If you have questions about the locations of libraries or other software, contact CISL Consulting Services at consult1@ucar.edu or 303-497-1249.

    Changing your NCAR password and shell

    NCAR users' password will be in sync with the UCAR Central Authentication Services (UCAS) password. If you desire to change your UCAR password, go to:

      http://www.ucar.edu/csac/ssh-auth-password-help.html

    Please see Password guidelines and instructions for changing passwords on NCAR computers.

    If you want to change your shell:

    1. Log on to frost
    2. chsh -s shell

    Setting up up your TeraGrid environment - CTSS

    Your TeraGrid software environment is set up to use the Coordinated TeraGrid Software and Services (CTSS) software stack.

    The following CTSSv4 kits are available in production:

    • Core
    • Remote Login
    • Remote Compute
    • Data Movement
    • Application Development and Runtime Support
    • Parallel Application
    • Science Workflow Support

    The following CTSSv4 kits will not be supported on this resource:

    • Data Management
    • Visualization Support

    SoftEnv is a tool that makes it easy for users to modify Unix shell environment variables as necessary to make software available. To simplify the set up of your environment, SoftEnv uses keys that represent individual software packages or suites of related packages. All CTSS software may be accessed using SoftEnv keys. Your default login software is determined by the SoftEnv keys listed in the .soft file in your home directory. To see your environment settings, type softenv. See

      http://www.teragrid.org/userinfo/jobs/softenv.php

    for information on how to use SoftEnv.

    Job queues and charging

    Charges for NCAR accounts

    There are currently no charges for frost for regular NCAR accounts.

    Charges for TeraGrid accounts

    The TeraGrid uses a charging unit called a Service Unit (SU). TeraGrid users with startup (DAC) accounts have a limited amount of resources for running interactive jobs and are not charged. MRAC (medium) or LRAC (large) allocation usage is charged in SUs.

    The charging formula for TeraGrid projects that run on frost is:

      queue factor * machine factor * allocated cpus * wallclock time in hours

    Currently the machine factor is one. The queue factors are all one.

    You can view your SU usage via the TeraGrid portal web site at

      https://portal.teragrid.org/gridsphere

    Log in to the TeraGrid portal using your portal login and password and go to the "My TeraGrid" tab.

    File transfer methods

    NCAR frost users

    It is possible to use scp/sftp to transfer files between frost and other computers without using key-based authentication. However, it is much more convenient to have your SSH public key installed to access and transfer files between frost and other machines. Following is the method for installing keys.

    1. If you have not created an SSH key on your machine you may create it by giving the command:
        ssh-keygen -t rsa

      You may elect not to add a passphrase; just hit enter when prompted for a passphrase. If you add a passphrase, then every invocation of ssh/scp will ask you to key in the passphrase. We strongly suggest creating SSH keys with passphrases and using ssh-agent to avoid repeated prompts.

    2. Now copy the contents of ~/.ssh/id_rsa.pub and append it to the file on frost ~/.ssh/authorized_keys.

      Make sure your ~/.ssh file is writable only by you.

    Once this is done you may transfer files from/to your machine to/from frost by initiating the transfers from your machine. Examples:

    While logged on your local machine:
    To transfer a file from your machine to frost:

      scp file your_user_id@frost.ncar.teragrid.org:/ptmp/$LOGNAME/file

    To get a file from frost to your machine:

      scp your_user_id@frost.ncar.teragrid.org:/ptmp/$LOGNAME/file file

    If your site accepts incoming scp requests to your machine, then you may install your ssh key generated on frost onto your your local machine and be able to initiate transfers from frost.

    For TeraGrid users

    If you want to transfer files between two TeraGrid resources, we recommend using GridFTP, as described at:

    http://www.teragrid.org/userinfo/data/gridftp.php

    If you want to transfer files between your machine and a TeraGrid machine, we recommend that you:

    1. install MyProxy and GSISSH (part of the Globus toolkit) on your machine (Linux/Unix/Mac and through Portal and Java based in windows)

      For details please consult http://www.teragrid.org/userinfo/access/tgsso_native.php

    2. get a single-signon proxy certificate using the myproxy-logon method described in the document above.

    To transfer files, use the following example commands:

    To transfer files from your desktop to frost:

      gsiscp file your_login@frost.ncar.teragrid.org:file

    To transfer files from frost to your machine:

      gsiscp your_login@frost.ncar.teragrid.org:file file

Troubleshooting program failures

If you experience an unexplained program failure, you can seek help from CISL Consulting Services. First, prepare a brief description of the problem and collect the following information into an email message:

  • Job number
  • Nodes on which the job was running, if known
  • Time the problem occurred
  • Location of the executable, run script, and error output

If you are an NCAR user, please send email to frost-help@ucar.edu. If you are a TeraGrid user, send email to help@teragrid.org.

Frost contacts and news

Contacts for NCAR users

  • User support: frost-help@ucar.edu
  • General Discussion: bglusers@ucar.edu

Please subscribe to the bglusers and bglstatus mailing lists via the following links:

BGLUsers: http://mailman.ucar.edu/mailman/listinfo/bglusers

    Blue Gene/L Supercomputer (frost) users should use the bglusers mail list to discuss application issues.

BGLStatus: http://mailman.ucar.edu/mailman/listinfo/bglstatus

    The bglstatus mail list is used to send email notifications regarding Blue Gene/L (frost) and the Front End Nodes uptime, downtime, emergency downtime, systems problems, etc.

Contacts for TeraGrid users

  • User support: help@teragrid.org
  • TeraGrid User and System News: http://news.teragrid.org
You may subscribe to get news postings about uptimes for frost and announcements for other TeraGrid systems at http://news.teragrid.org.

Next page | Table of contents - Frost user guide

If you have questions about this document, please contact CISL Customer Support. You can also reach us by telephone 24 hours a day, seven days a week at 303-497-1278. Additional contact methods: consult1@ucar.edu and during business hours in NCAR Mesa Lab Suite 39.

© Copyright 2004-2007. University Corporation for Atmospheric Research (UCAR). All Rights Reserved.

Address of this page: http://www.cisl.ucar.edu/docs/frost/overview.jsp