We are working to create information resources that you can use in order to optimize the use of the cluster for your research needs. Note: Some of the following tutorials may reference other clusters like University of Texas Stampede and SDSC Comet, but the information still applies to use the Hummingbird.
- Globus Data Transfer Tutorial on the usage of Globus with Science DMZ (Cornell Virtual Workshop)
- Note: Our Science DMZ is accessible only from the Data Center.
- Install Globus Client on Mac
- Install Globus Client on Windows
- Install Globus Client on Linux
- Learn Linux (Cornell Virtual Workshop)
- Linux/UNIX tutorial
- Python Introduction (Cornell Virtual Workshop)
- Python for high performance computing (Cornell Virtual Workshop
- R on HPC
- Lots of tutorials at HPC University
Documentation to get you started with SLURM and to gain more advance knowledge of future cluster usage:
- How to create scripts for SLURM to run your software/program
- How to use SLURM to execute your script file (University of Utah)
- SLURM Batch Scripting (University of Utah)
WARNING: Before implementing programs against the examples provided by these sites, make sure you have first reviewed “Start Here” page of this web site, “Using SLURM on Hummingbird“. Clusters all vary in the manner that the SLURM scheduler is used to manage resources. They also differ in the underlying hardware architecture in use and available. Therefore, the parameters and methods required may differ; different constraints and recommendations may apply.
Backup strategies
The following are links to various ways to backup and/or archive your data etc to cloud storage solutions.
Amazon tool awscli
S3 Sync Script s3cmd
Google drive Sync