The University of Iowa has partnered with several national computing and scientific organizations in order to provide additional computing resources to UI researchers and staff. Most notable are our connections with the Extreme Science and Engineering Discovery Environment (XSEDE), the Great Lakes Consortium for Petascale Computation (GLCPC), and the Blue Waters Project.
Xsede is a robust collection of integrated digital resources and services funded by NSF. This is a single virtual system that scientists and researchers can use to interactively share computing resources, data, and expertise. XSEDE integrates the resources and services, makes them easier to use, and helps more people use them. The five-year, $121 million project is supported by the National Science Foundation, and replaces and expands on the NSF TeraGrid project.
A part of the Xsede project is the Xsede Campus Champions. Through the Campus Champions program, the UI campus will have direct access to XSEDE and input to its staff, resource allocations for the use of campus researchers, and assistance in using those resources.
XSEDE Supercomputing Clusters of Interest:
Jetstream is a cloud computing environment designed to provide XSEDE researchers and students user-friendly, on-demand access to interactive computing and data analysis resources. Jetstream is implemented and supported by Indiana University, Texas Advanced Computing Center (TACC), and several additional partner institutions across the nation. Jetstream provides a library of virtual machines pre-configured to support scientific analysis and collaboration in a variety disciplines (e.g., biology, earth sciences, and social sciences, among others). Generic Linux-based VM configurations also are available, allowing software developers and researchers to create their own custom VMs to support specific applications or disciplines. Creators of custom VMs can save them for future reuse either as "private VMs" or share them with other Jetstream researchers via the "public" VM library.
UI Usage of Jetstream
One good example of leveraging Jetstream is to utilize it as a data analytics environment for classes. For example, Jupyter Notebook and RStudio are definitely the most widely used open-source data analytics frameworks by Python users and R users, respectively. They run not only on personal computers, but also on HPC (High Performance Computing) driven servers like Jetstream resources. If users connect to Jupyter Notebook and RStudio running on servers in Jetstream, they can benefit from the powerful computing resources in learning and doing their own data analytics.
Bridges is a multi-tiered, data-intensive high-performance computing system hosted by the Pittsburgh Supercomputing Center (PSC). Featuring regular, large, and extreme shared memory compute nodes, GPU nodes, database nodes, web server nodes, and data transfer nodes, Bridges is designed for extreme flexibility, functionality, and usability, and is well-suited for "Big Data" projects in the life sciences, the social sciences, and the humanities.
UI Usage of Bridges
Starting Fall 2017, ITS-Research Services will be helping faculty to teach courses (such as Big Data Analysis from Management Sciences and/or Electrical and Computer Engineering or high performance computing classes from Computer Science) on XSEDE supercomputing clusters. For the Fall of 2017, Big Data Management & Analytics course (MSCI:6110) and High Performance and Parallel Computing (CS:4700) will be taught using Bridges clusters.
We recently helped one of our faculty members, Dr. Xun Zhou, to submit an Educational Allocation Proposal to XSEDE for teaching his course entitled Big Data Management & Analytics (MSCI:6110). We requested an allocation of 112,000 Service Units (SUs) and were successful in receiving 66,000 SUs. Below is the link to the PDF document that we submitted. This can be used as a template to submit an Educational Allocation Proposal.
Educational Allocation Proposal.
Educational Allocation Proposal requirements can be found on XSEDE website here.
The Great Lakes Consortium for Petascale Computation is a collaboration among colleges, universities, national research laboratories, and other educational institutions. The consortium facilitates the widespread and effective use of petascale computing through the development of new computing software, applications, and technologies. The University of Iowa is a founding member of this group which qualifies our researchers for participation in allocations on the Blue Waters system.
The Blue Waters project, which is supported by the National Science foundation and the University of Illinois, will deliver a supercomputer capable of sustained performance of 1 petaflop on a range of real-world science and engineering applications. It is expected to be one of the most powerful supercomputers in the world.
Blue Waters will be composed of more than 235 Cray XE6 cabinets based on the recently announced AMD Opteron™ 6200 Series processor (formerly code-named "Interlagos") and more than 30 cabinets of a future version of the recently announced Cray XK6 supercomputer with NVIDIA® Tesla™ GPU computing capability.