The Center for Computational Research (CCR) at the University at Buffalo stands as a pivotal resource for researchers, educators, and businesses alike. We provide access to cutting-edge supercomputing facilities, advanced visualization tools, and a local research cloud infrastructure, all backed by a team of experts dedicated to accelerating your research endeavors.
It’s important to note that CCR is specifically focused on research computing resources at UB. For all other IT services and support at the University at Buffalo, please visit UBIT.
For those new to CCR and our diverse offerings, we encourage you to explore our comprehensive website: https://buffalo.edu/ccr.
To gain a quick understanding of CCR’s services and how we support the UB research community and industry partners, we recommend viewing this concise 30-minute overview presentation: https://buffalo.edu/ccr.
Getting Started with UB Learns and CCR Training
For all new users of CCR, we strongly recommend enrolling in our self-paced “Intro to CCR” training course available on Ub Learns at https://ublearns.buffalo.edu/d2l/le/discovery/view/course/285151. This course, accessible through UB Learns, is designed to guide you through the essential concepts of utilizing CCR’s high-performance computing environment. It includes modules and practical examples to ensure a smooth onboarding experience. Best of all, it’s free for all University at Buffalo faculty, staff, and students. External collaborators, RPCI personnel, and industry clients can find recorded video modules on our YouTube channel.
Starting in the Fall 2024 semester, completion of the “Intro to CCR” course on UB Learns will be a mandatory prerequisite for all undergraduate and graduate students requiring CCR access for their coursework. Further details can be found here.
Choosing the Right Computing Resources at CCR
Need assistance in determining the best resources for your research? Contact CCR Help for personalized guidance.
CCR offers a range of powerful computing resources to meet diverse research needs:
High Performance Computing (HPC)
Key features to consider:
- Our HPC resources are structured within a cluster environment, optimizing performance for computationally intensive tasks.
- Users submit batch jobs to a scheduler, which intelligently distributes workloads across servers (nodes) within the cluster.
- The UB-HPC cluster is available to both academic researchers and industry partners, fostering collaboration and innovation.
- CCR also manages dedicated clusters for faculty who have privately invested in their own computing nodes, providing tailored solutions.
- Accessing these resources is straightforward via a SSH client or the user-friendly OnDemand web portal, offering flexibility and convenience.
- For UB faculty research groups, access to the UB-HPC cluster is provided at no cost, removing financial barriers to advanced computing (student accounts require faculty sponsorship).
- Storage is shared across all clusters, and more information on storage solutions can be found here.
Research Cloud
Key features to consider:
- Our on-premise research cloud is ideal for projects that are not optimally suited for the traditional HPC environment.
- Common cloud applications include database hosting, website infrastructure, container orchestration, software development, proof-of-concept studies, and a wide array of non-HPC scientific applications.
- Cloud computing time is billed based on CPU hour consumption, and storage is available in 1TB increments.
- Detailed billing information is available here.
Note: Access to any CCR resource requires an active allocation. More details on obtaining access can be found here.
Selecting the Appropriate Resources: Determining the ideal resource can be complex due to the wide spectrum of research needs and disciplines we support. If the descriptions above do not fully clarify your best option, please reach out to CCR Help for a consultation.
To assist you in this process, we may inquire about specifics such as:
- The software applications you intend to use.
- Whether your software requires commercial licenses.
- The software’s ability to run non-interactively (controlled by a pre-prepared file rather than a graphical interface).
- Compatibility with the Linux operating system.
- Your estimated requirements for memory, processing time, computing power, accelerators, storage, and network bandwidth for typical jobs.
- The frequency with which you anticipate running these types of jobs.
We understand that you may not have all the answers readily available. Please provide as much detail as possible, and our expert technical support team will guide you in identifying the most suitable resources for your specific research demands.
In conclusion, UB Learns is your first step to effectively utilize the powerful resources at CCR. Whether you are interested in High Performance Computing or Research Cloud, understanding the fundamentals through UB Learns and contacting CCR Help will ensure you are well-equipped to advance your research goals.
