MERCED Cluster
MERCED (Multi-Environment Research Computer for Exploration and Discovery) Cluster is 1872 core Linux based high performance computing cluster supported by the National Science Foundation Award ACI-1429783. The MERCED cluster runs with the Linux RedHat operating system, a standard flavor of Linux, and employs the Slurm job scheduler and queueing system to manage job runs.
How much does it cost ?
Service | Unit | UC Rate (in USD) | External (Non-UC) Rate (in USD) |
MERCED Cluster Cycles | Core-hour* | 0.01 | 0.02 |
Enhanced User Support for Research Computing | Per Hour |
98.69 |
108.20 |
* A core-hour is a single compute core used for one hour (a core-hour) and 2G of RAM
Interested in MERCED Cluster?
Please visit our documentation page and learn more.
Ready to sign up for an account on MERCED Cluster?
Please visit the UC Merced OIT Service Catalog and request an account (PI status or PI approval required).
For more information on using MERCED cluster or other Research Computing clusters see the sidebar
Pinnacles cluster
The NSF-MRI funded Pinnacles cluster located in the server facility is available for all faculty projects at NO COST! The Pinnacles cluster runs with the Rocky (8.10) operating system, and employs the Slurm job scheduler and queueing system to manage job runs.The Pinnacles cluster is equipped with the latest generation Intel Xeon Gold 6330 CPUs and NVIDIA Tesla A100 v4 40GB HBM2 GPUs.
Please visit our documentation page and learn more.
Ready to sign up for an account on Pinnacles Cluster?
Please visit the UC Merced OIT Service Catalog and request an account (PI status or PI approval required).
CHASE-CI
Cognitive Hardware and Software Ecosystem, Community Infrastructure (CHASE-CI) is a cloud of hundreds of affordable Graphics Processing Units (GPUs), networked together with a variety of neural network machines to facilitate development of next generation cognitive computing (NSF Award #1730158). Led by UCSD, CHASE-CI allows researchers across all 10 UC campuses to conduct research into advanced machine learning techniques. If you are interested in learning more about CHASE-CI, please contact CIRT.
Additional Computational Resources
Some researchers may have needs that exceed our campus computing capabilities - they may need more machines to complete a task quickly or they may need more specialized types of computing nodes (e.g. GPU nodes). Please refer to the resources such as the San Diego Supercomputing Center (SDSC) and the national-level computing resources known as ACCESS. SDSC and ACCESS offer significant amounts of computing time to researchers interested in trying HPC at large scales.
Information about additional resources through ACCESS Program can be found here.
UC Merced pilot JupyterHub infrastructure is on-line now !
Please go to the website at https://ucmerced.2i2c.cloud.
All active UC Merced students, staff and faculty should have access to the pilot hub.
1. You can login through UC Merced SSO
2. You can use the default jupyterhub image or you can change the image using the configurator. You can also create your custom user image. There is a guide about this here
3. General information about the user environment here.
If you have any additional questions accessing hub, please open a general OIT research computing ticket