Using data-access nodes

Logging in | Using GLADE scratch | Getting files from HPSS | Transferring files

The data-access nodes are provided to enable authorized users to perform a number of file-management operations. These include:

The data-access nodes are not compute nodes and any tasks users run on them are run "at risk." They also are not intended for use as long-term storage.

Short, non-memory-intensive processes are allowed, including such tasks as text editing or running small scripts for transferring files. If any task consumes excessive resources (determined at the discretion of the CISL consultants or system administrators), a system administrator will kill the process or processes and you will be notified.

No usage charges are assessed for reading data.

If you do not have an allocation but need to use the nodes to get data from CISL storage resources, submit a Small Data Access Request.

Logging in

To use these resources, log in to and authenticate as described in our Authentication and security documentation.

Use ssh as follows if your username on your local computer is the same as your UCAR username:


If your UCAR username and your username on your local computer are different, log in with your UCAR username as follows:


Using GLADE scratch

When you log in, you are in /glade/u/home/username by default. Use your larger scratch or work file spaces to hold data that you read out of HPSS. You can also read files from other GLADE file spaces to which you have been granted read permissions.

See GLADE for more information about file spaces.

Getting files from HPSS

After logging in, use either the HSI interface or the HTAR utility to fetch files from the High Performance Storage System (HPSS) archive.

This assumes you have an HPSS account and the necessary allocation and file/directory permissions. See this HPSS documentation for details.

Transferring files

To transfer data between and another system, use any of these tools:

See those links for documentation.