Computing
UC San Diego researchers have access to numerous compute platforms to meet the unique needs of every research project. From just-beyond-the-desktop to multi-core, multi-hour compute runtime requirements, there are small clusters and national resources, local hardware and cloud solutions to right-fit projects of every size and duration.
Cluster Computing
UC San Diego provides cluster computing resources for researchers to process data and perform large scale computations by distributing the computation workload across different nodes. In addition to the following resources, schools and / or departments may have local clusters available.
UC San Diego Research Cluster
UC San Diego Research Cluster
The UC San Diego Research Cluster provides no-cost access to CPU / GPU compute resources to graduate, post-doctoral, and faculty researchers for non-instructional, data-intensive projects.
The Research Cluster is used as a 'step-up' for projects needing more computing power than a personal workstation but less than that offered by a high-performance computing (HPC) resource. It is a customizable, containerized environment with two options for running jobs:
- Terminal / SSH interface: To run complex machine learning workflows, background batch jobs, and to utilize a full Linux/Ubuntu CUDA development suite. Library packages may be installed (e.g. conda/pip, CRAN) or one may opt to replace the default environment entirely by launching their own custom Docker containers.
- Web interface: A popular option for utilizing applications such as Jupyter Notebooks for combining live code, visualizations, text notes, and other code editor features. This offers straightforward interactive access to popular languages and GPU-enabled frameworks such as Python, R, Pandas, PyTorch, TensorFlow, Keras, NLTK, and AllenNLP. Learn more…
Triton Shared Computing Cluster (TSCC)
Triton Shared Computing Cluster (TSCC)
The Triton Shared Computing Cluster (TSCC) provides central colocation and systems administration for your cluster nodes (“condo cluster”), as well as providing a hotel service for those with temporary or bursty high-performance computing (HPC) needs. Learn more…
High-Performance Computing (HPC)
High-performance compute resources can be utilized for data intensive projects, to solve complex and large scale computational problems.
HPC@UC
The HPC@UC initiative provides the UC community with significant resources to aid their research. Awards granted under this program assist UC researchers who are inexperienced in HPC to expand their research program. Learn more…
ACCESS (Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support)
ACCESS is a national HPC ecosystem (a follow-on project to XSEDE), funded by NSF but available to research projects no matter the funding source. ACCESS resources include SDSC’s Expanse, TACC’s Stampede 3, PSC’s Bridges 2 and others. Learn more…
High-Throughput Computing (HTC)
Specializing in parallel computing, these HTC national resources are available to conduct research and process large volumes of data quickly and efficiently.
Open Science Grid (OSG)
The Open Science Grid Consortium (OSG) operates a fabric of international HTC services. Member institutions of the OSG Compute Federation contribute resources to this distributed system. OSG is a member of the ACCESS consortium. Learn more…
National Research Platform (NRP)
The National Research Platform (NRP, formerly PRP), hosts Nautilus, a ~300-node distributed compute cluster with CPUs, GPUs, and FPGAs designed as a testbed for science drivers to expedite science and enable transformative research. Learn more…
Cloud Computing Services
In addition to private cloud offerings, UC San Diego has agreements with commercial cloud providers that include University of California and UC San Diego specific discounts and other services.
Commercial Cloud
UC San Diego faculty, staff, and other affiliated researchers can access computing resources from cloud hosting services (Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP)). Learn more…
Amazon Lightsail for Research
Access research-class computing power along with pre-installed research software (RStudio, Jupyter, Ubuntu, VSCodium, Scilab), reducing time to research. Learn more …
AWS for Research Day
If you attended AWS for Research Day on 11/21/24, visit this link for details needed to get started with AWS.
SDSC Cloud
With the goal of providing a complete Data as a Service (DaaS) package, SDSC Cloud allows users to spin up instances on demand and can be paired with SDSC Cloud Storage for convenient, comprehensive big data analysis. Learn more...
Cloud Bank
UC San Diego, with University of Washington and University of California, Berkeley, supports the NSF-funded Cloud Bank which allows use of public cloud for research and education. Learn more...
UC San Diego Colocation Facilities
If you choose to purchase your own rack-mount compute equipment, UC San Diego has campus colocation data centers to house your equipment efficiently and securely.
San Diego Supercomputer Center (SDSC)
The SDSC facility is especially well suited to installations that must demonstrate regulatory compliance; that require monitoring and high uptime; that require backup power; and large installations. Learn more…
Scripps Institution of Oceanography (SIO)
Scripps Institution of Oceanography (SIO) offers free colocation services to the SIO community. Learn more…