Feel free to copy relevant portions of this text into your research proposals. This text includes language which is appropriate for inclusion in facilities descriptions for NSF, NIH, and other agency proposals.
The FSU Research Computing Center (RCC) operates as an academic services unit within Information Technology Services (ITS). The RCC Director oversees nine permanent professional staff and a variable number of term-limited project staff and students.
The RCC staff and students are responsible for maintaining core systems and are assigned to work in support of specific research domain projects. The RCC staff offices are located on the main campus of FSU in the first floor of the Dirac Science Library.
High-Performance Computing
The FSU HPC system is comprised of 17,600 x86 64-bit compute cores linked together by low-latency InfiniBand networks for MPI communication. The aggregate peak performance of the system is 512.8 TFLOPS. Compute nodes support between 4 and 64GB of memory per core, with servers up to 256GB. A redundant cluster of specialized login nodes serves as the user entry point for the system.
The cluster utilizes the SchedMD Slurm scheduler for job submission and scheduling. Slurm is an open-source scheduling system that automates the job submission, allocation, and management process for HPC users. To maximize the utility of the system, we offer a broad set of compilers, math/communication libraries, and software applications are available to users. Users can request the installation of new software by RCC staff. Using container technologies such as Docker and Singularity, users are free to configure their software environments for individual research projects.
Interactive Computing and Scientific Visualization
Large datasets generated on the RCC computing resources or by research instrumentation on or off FSU's campus can be interactively analyzed and explored using the "Spear" Interactive Computing Cluster. The Spear Cluster is comprised of over 280 x86 64-bit processor cores that are linked with other RCC resources over a high-speed IP network. Users log directly into Spear nodes and run interact with data through a basic shell or X11-based graphical applications.
Data Storage
The RCC maintains two file systems to facilitate data analysis pipelines and workflows.
- A 3.1 PB parallel file system (GPFS) is mounted on all of the HPC compute, login, and storage export nodes over a 40Gbps network.
- A 1.41 PB archival storage system is available through Globus GridFTP for long-term storage and sharing of research data.
Researchers can transfer data to and from these storage systems via the highly reliable and performant Globus Service. Users can also transfer data via more common protocols, such as SFTP or rsync.
Data Center Facilities and Network Connectivity
Computing and data storage resources managed by the RCC occupy reside in a managed data center located in Innovation Park in the Bernard Sliger Building.
The facility is equipped with raised floors and redundant HVAC cooling systems, extensive power distribution systems, large format UPS battery backup systems, and diesel-powered backup generators for prolonged outages. The data center is connected to the campus enterprise and research networks via multiple 40Gbps and 100Gpbs connections.
The data center also connects directly to the Florida LamdaRail (FLR), a dedicated 10Gbps regional optical network. The FLR makes high-speed data transfer and collaborative storage possible among major Florida universities.
Florida Cyber Infrastructure
Florida State University is a founding member of the Sunshine State Education and Research Computing Alliance (SSERCA), which was created in 2010 to bring together Florida's geographically distributed educational institutions in a way that maximizes their collective impact on research and education. SSERCA provides the management and technology framework to share and access resources distributed across the State of Florida.
The alliance currently supports several projects with sophisticated workflows and complicated data and compute requirements. More details regarding how SSERCA is accelerating research in the State of Florida are available at http://sserca.org. Current member organizations include FAMU, FIT, FIU, FSU, UCF, UFL, UM, UNF, and USF. The Florida LambdaRail provides connectivity to all institutions in the alliance.
Research Collaboration and Consulting
The RCC staff collaborates and supports grant-funded research in a large number of disciplines at Florida State University. Working collaboratively with researchers, RCC staff develop data processing workflows, deploy systems, install and maintain software, and advise on the appropriate use of technology. RCC staff participate in the development of custom software solutions and implementations of existing tools and resources to enable research objectives in different scientific domains at FSU.
Examples of such applications include database deployments, custom web apps, data processing pipelines, and data collection instruments.
Education
The RCC supports its mission by providing educational opportunities to the research community at FSU. Educational initiatives at the RCC include public workshops and seminars, online guides and documentation, and one-on-one engagement with faculty and students. The objectives of this effort are to increase the efficacy of research in the University community and to decrease barriers for researchers that may benefit from using RCC systems and services.
Cloud Computing and Partner Services
The RCC collaborates with other units in FSU Information Technology Services to provide custom virtual servers to research groups that need custom workflows. This resource enables quick deployment of web, database, and other non-high-performance computing applications. Servers in this system can connect to and leverage the additional storage and compute resources provided at the RCC.
The RCC also collaborates with the Open Science Grid (https://www.opensciencegrid.org/) and ACCESS (https://access-ci.org/ -- formerly XSEDE) to make additional resources available to the FSU Research Community.