You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 6 Next »

The SLAC theory group currently has 2 computing nodes that can be used by any member of the theory group.
The nodes are maintained by the SLAC IT and provide a linux infrastructure.

Access

In order to gain access to the machines your SLAC user account must be added to the user group "theorygrp".
If you cannot log in yet and you are not part of this group, send a ticket to SLAC IT and request to be added to the group in order to access the machines.

Currently, also Alex holds administrative rights on the machines.

Address

Login

Access via ssh is possible from within the SLAC network. To connect from outside, use a VPN.
To connect from a terminal use for example

    ssh username@epp-theory01.slac.stanford.edu

Alternatively, ssh into one of the SLAC gateway machines. These are accessible without VPN. For example,

    ssh username@iris.slac.stanford.edu
ssh username@rhel6-64.slac.stanford.edu

Once there, ssh to the theory machines,

    ssh epp-theory01
ssh epp-theory02

More details are available here.

Storage Space

In addition to your AFS home directory, we have local drives in our theory machines.
This local drives are mounted on the machines at

  • /nfs/farm/g/theory/u1
  • /nfs/farm/g/theory/u2
  • /nfs/farm/g/theory/u3

and provide 22TB, 14TB and 27TB of disk space respectively.
Make sure that enough space is available for your files when writing to these drives, for example by using

    df -h

When you log out of the computing nodes, your writing privileges on AFS will be revoked but you will still be able to write onto these disks.

Software

Special software can be installed globally on our machines - contact SLAC IT.

Many standard tools are already available via the SLAC system - see here.

Usage

The machines are used in a shared fashion - please respect the computing needs of others that might be using them at the same time.
Check how much memory and CPU capacity is available for example by using the command

    htop

To run programs for extended periods of time or when disconnected from the computing nodes use programs like screen.

  • No labels