Facilities Summary for Grant Proposals

The below summary can be used as a starting point for a "Facilities, Equipment and Other Resources" section for an NSF proposal.

We strongly encourage you contact researchit@iastate.edu to discuss the details of your project so we can help craft an appropriate list of applicable equipment and determine if additional equipment, storage, or other resources such as hosting costs, consulting, or developer time should be included in your proposal.

Human Resources

The ResearchIT team specializes in collaborating with researchers to solve complex technology problems, freeing them up to focus on their research. We support several large clusters, research server grids, large scale research storage, virtual server and container based environments for research, web applications, and hundreds of scientific software applications.  The team is composed of 5 full time staff and partial time from a number of other Systems Analysts, with a total combined FTE of approximately 7 specialists who provide their expertise to consult on projects and manage the various services that we offer.

Equipment and Facilities

Iowa State, through the help of the NSF MRI program and other funding programs and agencies, has invested heavily in computational resources.  We have several generations of University wide clusters available for use (detailed below).  In addition to our campus HPC clusters we also have more than two dozen specialized servers for performing diverse tasks such as large memory genome assembly, many core parallel jobs such as alignment, high throughput single core jobs, and machine learning using the latest GPUs.  We have a robust virtual machine and container based infrastructure with enterprise-grade performance and availability that allows us to host web applications with excellent up-time.  We also have a multi-petabyte storage system for secure retention of long term research data.  All of our equipment is professionally managed by IT staff, and housed in our secure, climate controlled data center.

Equipment summary:

  • Pronto Cluster:
    • GPU (20 nodes): 5x 1080Ti, 14x v100, 20x Quadro RTX 6000, 4x 2080Ti, 40x A100, 4x A40
    • Swift (11 nodes): AMD servers with 64 cores and 512 GB RAM
    • Speedy (12 nodes): high clock speed servers with dual processors (for high throughput jobs)
    • Biocrunch (12 nodes): each 176 threads, 1.5TB RAM (servers for shared memory multi-threaded jobs)
    • Bigram (2 nodes): one with 3TB RAM & one with 4TB RAM
    • Legion (8 nodes): KNL nodes, each 272 threads, 384G RAM
  • Nova Cluster: Approx 200 nodes, each at least 36 cores, 36% 192G, 34% 384G RAM, 27% 512 GB, high memory node, and GPU nodes with 9 total V100 GPUs & 60 A100 GPUs.
  • Virtual servers: a cluster of over twenty hypervisors available supporting VMs for development, testing, and web application hosting
  • Container environment: a Kubernetes-based container hosting environment for research applications
  • Storage: a multi petabyte long term storage system (LSS), and approx 1 PB of high-performance scratch space

For more detailed equipment information, please see: https://rit.las.iastate.edu/hardware