Skip Nav

Sapphire




Cray XT3 (sapphire)
Beginner's FAQs




Q. How do I logon to Sapphire from a Windows or Unix client?
A. For logon instructions from a Windows client, click here. For a Unix client, click here.

table of contents  —   top of page

Q. Which protocol should I use to connect to Sapphire?
A. There are basically three protocols supported for login access to our systems. The protocols are telnet, rlogin , and ssh. We recommend that you use ssh. Below is a table containing the pros and cons of each protocol.

ProtocolProsCons
ssh Encrypted sessions allow tunneling of X11 connections to reduce firewall problems and are more secure because they are encrypted. Can be tricky to configure
rlogin Encrypted sessions No tunneling
telnet Perhaps most familiar No encryption; no tunneling

table of contents  —   top of page

Q. Why do I have to keep getting new tickets?
A. A Kerberos ticket has a limited lifetime. After the ticket has expired, you must get a new ticket before you can open a new terminal session, start a file transfer, or use an rsh command. The length of the ticket's life is dependent on your location.

For a list of the current Kerberos ticket life expectancies, see the HPCMP Ticket Lifetime Matrix External Link. Note: You will be required to authenticate using Kerberos before viewing the page. Users with browsers other than Internet Explorer may have difficulty authenticating.

table of contents  —   top of page

Q. How do I generate a batch submission script?
A. The Portable Batch System (PBS) is the batch system for Sapphire. The
User Guide for Sapphire has a sample PBS script. To submit a batch job, use the following command:

qsub scriptname

where scriptname is the name of the file containing the batch script. For more information, see the PBS qsub Quick Reference Guide. Typically, each batch job needs a valid executable, a submission batch script, and any input files or data that are required. Usually the job is run in the user's $WORKDIR or $WORKDIR2 , which are large scratch disk spaces. The executable can either be from a program that the user has compiled or from one of the software packages that are in a shared disk space.

Your job script must contain a subproject ID against which the run time is to be charged. Use the show_usage command to generate a list of your current subprojects. For more information on qsub or show_usage, see their respective man pages.

table of contents  —   top of page

Q. How do I get started running batch jobs?
A. Production jobs are submitted and scheduled in a batch queuing system utilizing the PBS scheduler. Typically, each batch job needs a valid executable, a submission batch script, and any input files or data that are required. Usually the job is run in the user's $WORKDIR or $WORKDIR2 , which are large scratch disk spaces. The executable can either be from a program that the user has compiled or from one of the software packages that are in a shared disk space.

A list of available software and locations for Sapphire can be found at http://www.erdc.hpc.mil/documentation/XT3_Docs/software.

Information on the available compilers and typical compiler flags for Sapphire can be found in the "Program Development" section of the XT3 User Guide.

table of contents  —   top of page

Q. How do I run an interactive job?
A. In order to use the interactive batch environment, you must first acquire an interactive batch shell. This is done by executing a qsub command with the "-I" option from within the interactive environment. For example,

qsub -l ncpus=# -A project_name -q queue_name -l walltime=wall_time -I

Your batch shell request will be placed in the desired queue and scheduled for execution. This may take a few minutes because of the system load. Once your shell starts, you will be logged in to one of the PBS host nodes. At this point, you can run or debug interactive applications, execute job scripts, start an execution on the compute node via the aprun command or postprocess data, etc.

table of contents  —   top of page

Q. Where do I put my executable and output files?
A. $HOME is the recommended location for building executables. The executable should then be copied to and run from $WORKDIR or $WORKDIR2. Output files should also be written to $WORKDIR or $WORKDIR2.

Files in your $WORKDIR or $WORKDIR2 that go unused for more than a few days may be automatically deleted, especially if /work or /work2 becomes full.

If you use $WORKDIR or $WORKDIR2 in your batch scripts, you must be careful to avoid having one job accidentally contaminate the files of another job. If two different batch jobs use the same names for temporary files, unusual errors can arise if the two jobs happen to run at the same time. By having each job create and use its own subdirectory underneath $WORKDIR or $WORKDIR2, this problem can be avoided.

table of contents  —   top of page

Q. How do I tell what version of Kerberos client binaries I am using?
A. For Unix/Linux clients, the best way to tell is to issue the following command:

strings /usr/local/krb5/bin/rlogin | grep NRL_RELEASE_

For Windows, you need to look at the three applications (krb5.exe, putty.exe, filezilla.exe). Launch the krb5.exe executable. Click on "Help->About Kerberos". For PuTTY, launch the executable and click on "About". For FileZilla, click on "Help->About Filezilla".

table of contents  —   top of page

Q. How do I tell the latest required version of the Kerberos client binaries?
A. For a list of required versions for all of Kerberos binaries (Windows and Unix), see the HPCMP Software Minimum Required Versions Table External Link.
Note: You will be required to authenticate using Kerberos before viewing the page. Users with browsers other than Internet Explorer may have difficulty authenticating.

table of contents  —   top of page

Q. What is Information Environment (IE), and what does it have to do with getting access to Sapphire?
A. IE is a Web-based system maintained by the HPCMPO that is used by Service Agency Approval Authorities (S/AAAs) to manage and measure usage of HPCMP resources. For more information, please visit the Information Environment link, located in the Community section of the HPCMPO External Link Web site.

table of contents  —   top of page

Last update: January 22, 2010

You are accessing a U.S. Government (USG) Information System (IS) that is provided for USG-authorized use only. By using this IS (which includes any device attached to this IS), you consent to the following conditions: * The USG routinely intercepts and monitors communications on this IS for purposes including, but not limited to, penetration testing, COMSEC monitoring, network operations and defense, personnel misconduct (PM), law enforcement (LE), and counterintelligence (CI) investigations. * At any time, the USG may inspect and seize data stored on this IS. * Communications using, or data stored on, this IS are not private, are subject to routine monitoring, interception, and search, and may be disclosed or used for any USG- authorized purpose. * This IS includes security measures (e.g., authentication and access controls) to protect USG interests--not for your personal benefit or privacy. * Not withstanding the above, using this IS does not constitute consent to PM, LE or CI investigative searching or monitoring of the content of privileged communications, or work product, related to personal representation or services by attorneys, psychotherapists, or clergy, and their assistants. Such communications and work product are private and confidential.