HPC Cluster

The Miami Redhawk cluster (currently 360 cores) is available for use by faculty, staff, and students.  Contact the RCS group to arrange a discussion about how the cluster can support your research and teaching efforts.

For larger computing needs, the RCS group can assist researchers with using resources available from the Ohio Supercomputer Center.

Accounts

Faculty can activate their cluster account by visiting the Cluster Account Management web site and logging in with their Miami ID and password.  A successful login will trigger the account creation process, and the cluster account will be active within 20 minutes.

Student and staff accounts need to be sponsored by a faculty member.  This can be done using the “Add User” button on the  Cluster Account Management web site.

Access

There are different ways to access the Redhawk cluster.  Note that when you are accessing the cluster, you are connecting to the head or login node of the cluster.  See the section on using the cluster for information about accessing the compute nodes.

Command line access is available using SSH, with SFTP/SCP used to transfer files.  See the following documentation for more details:

Access to the cluster with a full Linux desktop is also available using a tool called NoMachine or NX.  File transfers will still need to use SFTP/SCP – see the documentation above for information on transferring files.  The following documentation contains details on using NoMachine (note that a Linux client for NoMachine is also available):

Hardware

Current details about the cluster hardware can be found on the HPC Hardware page.

Software

Most software on the cluster is managed with the modules tool.  Details about software currently installed on the cluster can be found on the HPC Software page.

Using the Cluster

Cluster usage is broken into two categories – interactive and batch.  Interactive use allows the user to interact with a cluster node to run software and receive output in real-time.  In batch usage, work is submitted to cluster and executes when the needed resources are available, with optional e-mail notifications to the user when the job starts and ends.  More details can be found at: