What is U2?

• U2 is the major computational resource in CCR. It is a cluster of Linux machines used for scientific computations (> 2000 processors).
• The U2 page provides details on this cluster, including a schematic.

Where can I get help?

• Guides for new and experienced users are available on the Getting Started page.
• The Training page has a list of online tutorials and workshops.
• Users can ask the CCR staff questions. The Getting Help for CCR Resources page has the details.
• The CCR staff frequently consult with researchers on a wide variety of topics. The Consulting page describes the CCR consulting services.

How can I get an account?

• UB faculty can apply for accounts using a form on the Accounts page.

How do I login?

• Logins are allowed from the UB domain using ssh.
• The Login to U2 page has instructions on connecting to CCR from Linux/Unix and Windows machines.
• A step-by-step guide for new users can be found on the Getting Started page.

How do I logout?

• On the command line, type logout or exit.

How do I change my password?

• On u2: A password is changed by using the kpasswd command.
  • kpasswd
• On lennon: A password is changed by using the yppasswd command.
  • yppasswd

NOTE: If you can not remember your password, you can reset it for most CCR resources in MyCCR. This does not apply to lennon - please contact CCR help to request a password reset on lennon.

How do I display a graphical interface from CCR to my workstation?

• This requires logging in to U2 with X-Display enabled.
• The X-Display pages describes how to set the X-Display on Linux/Unix and Windows machines.

How do I transfer my data files?

• The Transferring Data page provides information on uploading and downloading data files.

Where are my files?

• Files are in the user's home directory by default. The home directories are located on the SAN (Storage Area Network) and are accessible from all compute nodes in the U2 cluster.
• Research groups may have a shared projects directory. Project directories are accessible from all compute nodes.
• Users may also use temporary space. This is referred to as scratch space, and is intended for temporary use, such as while running programs. There are three different scratch spaces:
  • global scratch (/san/scratch)
  • high performance global scratch (/ibrix/scratch)
  • local scratch (/scratch on each compute node)

Are my data files backed up?

• User data are backed up according the CCR Backup and Retention Policy.

What is my quota?

• The default quota are 2GB for user home directories.
• Projects directories have a default group quota of 100GB.
• To find out your quota and usage, use the command: quota -v

What software packages are available?

• The Application Software page is a good place to start. This page list all the current scientific and visualization applications available on the U2 cluster, as well as compilers, mathematical libraries and MPI implementations.
• Users can request software installations by contacting the CCR staff.

How do I run a job?

• CCR uses the PBS batch scheduler to run jobs.
• Details on submitting and runnings jobs can be found on the Running Jobs page.

When will my job run?

• The showstart command will give an estimated start time of job.
  • showstart jobid
  • The Running Jobs page has a list of commands that provide information on job and queue status.

What factors determine the priority of a job?

• There are several factors that contribution to the calculation of the priority. These include number of compute nodes requested, time waiting in the queue, contribution level, and the user's fairshare priority among other factors.
• The CCR Job Priority page has the details on the job priority.

How do I find out more about PBS?

• PBS Users Guide provides instructions, examples and screen-shots of using the PBS/Torque scheduler.