Campus Research Computing Options
Research computing is complex and multi-faceted, incorporating a variety of resources and models. These can include high-performance computing, stand-alone computation systems, cloud computing, and more. No self-help guide or IT professional can tell you what resources you'll need to satisfy your computing requirements. Our goal with this document is to help you to narrow the scope of items that may be helpful to your research by eliminating those that will NOT be helpful.
The Grainger College of Engineering investment in this resource is available for both research and instructional purposes (currently totaling 102 nodes with 2,636 CPU cores and 20 GPUs). Additionally, Illinois has a 4,300-core High-Throughput Computing resource available based on HT Condor. Both of these resources can be used at zero cost to anyone affiliated with Grainger Engineering.
The chart below compares a small sample of research computing services at Illinois. A more complete list can be found through the Research IT Portal.
Click image to see enlarged version
If none of the computing resources listed above meet your needs, Engineering IT is available to help create and support a solution tailored to your project’s computing requirements. Email engrit-help@illinois.edu to get started.
The Grainger College of Engineering investment in the Illinois Campus Cluster Project includes:
- 102 infiniband interconnected compute nodes providing 2,636 CPU cores and 20 NVIDIA GPUs, comprised of:
- 8 CPU nodes - 2x Intel Xeon E5-2680 v4 (14-core), 224 cores, 2048 GB RAM (256 GB ea.)
- 76 CPU nodes - 2x Intel Xeon E5-2690 v3 (12-core), 1512 cores, 19,456 GB RAM (256 GB ea.)
- 7 GPU nodes, 2x Intel Xeon Gold 6148 (20-core), 280 cores, 1344 GB RAM (192 GB ea.), 14x V100 GPUs
- 3 GPU nodes, 2x Intel Xeon E5-2680 v4 (14-core), 84 cores, 768 GB RAM (192 GB ea.), 6x P100 GPUs
- 8 CPU nodes, 2x Intel Xeon E5-2680 v4 (14-core), 224 cores, 512 GB RAM (64 GB ea.)
- 5TB of shared storage
NOTE: As part of this investment, the 8 CPU nodes with 64 GB of RAM and the 3 GPU nodes with P100 GPUs are dedicated to instructional use.
Training Resources
- The University of Chicago has put together a brief glossary of terms used in research computing.
- If you are new to Illinois Campus Cluster, these video tutorials may help you get started.
- If you are new to research computing in general, HPC University offers an Introduction to HPC and Supercomputing and many other courses.
- For hands-on workshops, Computational Science and Engineering offers training in a variety of areas to help you make the most of whatever resources you use.
- Additional resources can be found at Lynda.com (free to University of Illinois users, but login first here).