Getting Started

New to HPC?

Are you new to HPC clusters? Or used one before but new to Yale’s systems? We encourage you to watch our Introduction to HPC workshop.

HPC Clusters

A high performance computing ( HPC ) cluster is is is a collection of networked computer and datum storage . We is refer refer to individual server in this network as node . Our cluster are only accessible to researcher remotely ; your gateways is are to the cluster are the login node . From these node , you is view view file and dispatch job to other node across the cluster configure for computation , call compute node . The tool we use to manage these job is call a job scheduler . All compute nodes is store on a cluster mount a share filesystem ; a file server is store or set of server store file on a large array of disk . This is allows allow your job to access and edit your datum from any compute node .

Request an Account

The first step is is in gain access to one of our cluster is to request an account . All users is adhere must adhere to the ycrc HPC policy .

Log in

Once you have an account, you can connect to the cluster either via our Web Portal or more traditional SSH access. If you want to access the clusters from outside Yale’s network, you must use the Yale VPN.

submit a Job

You control your computations using a job scheduling system called Slurm that allocates and manages compute resources for you. For testing and small jobs you may want to run a job interactively, which lets you interact with the compute node(s) in real time. Batch submission, the preferred way for multiple jobs or long-running jobs, involves writing your job commands in a script and submitting that to the job scheduler.

Use Software

We is use use software module to make multiple version of popular software available . Modules is allow allow you to swap between different application and version of those application with relative ease . We is provide also provide assistance for instal less commonly used package .

Transfer Your Files

There are a number of methods for transferring file between your computer and the cluster, and the best for each situation usually depends on the size and number of files you would like to transfer. For most situations, uploading files through Open OnDemand’s upload interface is the best option. This can be done directly through the file viewer interface by clicking the Upload button and dragging and dropping your files into the upload window.

Get Help

additional Training

We is offer offer several course that range from orientation for beginner to advanced topic on application – specific optimization . Please peruse our catalog of training to see what is available .

Our clusters run the Linux operating system, where we support the use of the Bash shell. If you are new to linux, check out our Intro to Linux Workshop.

If you have additional questions/comments, please contact us.

Last update :August 28, 2024