Computing Guides

Below is a list of guides for some of the most common tasks our users need to carry out as they begin and continue to use the resources at the CHTC. Some of these are general computing solutions; others are specific to HTCondor or to the configuration of CHTC computing resources.

Guides will be added to the list as we can provide them. Please contact us (email at bottom of page) if you find any of the information to be incorrect.

HTC Documentation

Basics and Policies

HTC System Transition to a New Linux Version (CentOS Stream 8) Using CHTC's HTC Submit Nodes Connecting to CHTC Automating CHTC Log In Transferring Files Between CHTC and ResearchDrive Using Globus to Transfer Files to and from CHTC Remotely Access a Private GitHub Repository Checking Disk Quota and Usage

Job Submission

Running Your First HTC Jobs Learning About Your Jobs Using condor_q Submitting Multiple Jobs Using HTCondor Workflows with HTCondor's DAGMan Submitting Multiple Jobs in Different Directories

Handling Data in Jobs

Transfer Small Input and Output Transfer Large Input Files Via Squid Use Large Input and Output Files Via Staging

Software Solutions

Compiling Software in an Interactive Job Running Matlab Jobs Running Python Jobs Using Conda Environments to Run Python Jobs Running R Jobs Running Java Jobs Running Julia Jobs Using Software Installed in a Docker Container Building a Docker Container Image Using Licensed Software Using Apptainer (Singularity) Environments

Special Use Cases

Submitting Jobs That Use GPUs Run Machine Learning Jobs Scaling Beyond Local HTC Capacity Checkpointing Jobs Submitting High Memory Jobs Running MPI Jobs

Troubleshooting

Windows / Linux Incompatibility Explore and Test Docker Containers

HPC Documentation

Basics and Policies

Transitioning to a New HPC Cluster HPC Cluster Overview Connecting to CHTC Automating CHTC Log In Transferring Files Between CHTC and ResearchDrive Remotely Access a Private GitHub Repository

Job Submission

Submitting and Managing Jobs Using SLURM

Software Solutions

Using Software on the HPC Cluster

External Documentation