Skip navigation and jump directly to page content

 IU Trident Indiana University

UITS Research Technologies

Cyberinfrastructure enabling research and creative activities

About Big Red and IU’s advanced cyberinfrastructure

  • Researchers using Big Red have led IU in securing more than in $253 million in grant funding since 2006.
  • Big Red is now considered old by supercomputer standards. When it retires in early 2013, Big Red will have been IU’s main supercomputer for nearly seven years. Normally, a supercomputing system stays in operation only four or five years. We are at a time when we have to replace Big Red, or risk a hardware failure that might disrupt critical IU research.
  • Big Red II will be unique among university-owned supercomputers, given its one petaFLOPS peak operating rate, advanced internal network, and capabilities for handling big data challenges.
  • Over the past 15 years, IU has made significant investments in its high-performance computing resources, resulting in a number of “firsts.” In 2001, IU’s Research SP was the first university-owned supercomputer capable of one teraFLOPS processing capability – a trillion mathematical operations per second. In 2003, IU claimed the first distributed Linux cluster achieving more than one teraFLOPS on standard benchmark applications. In 2006, Big Red was the fastest academic supercomputer in the western hemisphere.
  • Big Red II will be more than 25 times faster than the original Big Red. Big Red II, which will have more than 21,000 computer processor cores, will succeed Big Red, which has 4,096. Since joining IU’s computing arsenal, Big Red has run more than 3 million computing jobs, using more than 125 million hours of computer processor time. Its speed has enabled research in medicine, biology, bioinformatics, chemistry, astronomy, physics, geography, climate studies, informatics, network science, sociology, and public health.
  • With an internal network that is advanced, fast, and high-bandwidth, Big Red II will open up new research possibilities – particularly in “Big Data” – that IU was simply not able to support before. Some conversion work will be necessary for those well served by Big Red and its predecessors, but Big Red II has a “cluster compatibility mode” and is based on AMD processors that are compatible with commodity Intel processors (such as those in the Quarry Cluster).
  • Quarry will remain IU’s primary parallel computing cluster. While much smaller than Big Red II, Quarry will serve as a ‘condominium computing’ environment for IU, allowing continued consolidation of IU departmentally owned computing systems into the highly secure and energy efficient IU Bloomington Data Center. Condominium computing lets departments use grant and departmental funding to purchase computer nodes and install them as part of the Quarry cluster. This has several benefits: professional management, backups, and security; availability within minutes; expanded computing capability for the entire IU community; and conservation of natural resources and energy.
  • Big Red II is an investment that will pay for itself many times over in terms of economic development in Indiana. IU researchers who use Big Red have secured grant funding that totals more than 25 times what it cost to build Big Red. We expect similar or greater returns on IU’s investment in Big Red II.
  • Since IU purchased Big Red, two high-tech companies and a new venture capital company have opened offices in Bloomington.  IU supercomputers, paired with the expertise of the School of Informatics and Computing, are the base on which this local economy grows. In the long run, one of the best things IU can do is to help the Indiana economy grow through innovative uses of advanced technology and adaptations to 21st century global economic competition.
  • There are a few bigger supercomputers than Big Red II, but none of them are owned and operated by an American university. The Blue Waters system, which is being installed at the University of Illinois at Urbana-Champaign, will be about 10 times bigger – but a national review panel will allocate it, highly restricting its use. At IU, any member of the university community involved in research – from a freshman to a distinguished professor – will receive access to Big Red II. 
  • FLOPS, or floating-point operations per second, are a measure of computing performance. As a one petaFLOPS peak supercomputer, Big Red II will process a thousand trillion operations per second. If you performed one calculation per second with a calculator, it would take you 31.7 million years to do what Big Red II can do in a second.
  • While cloud computing will have a role in IU’s research computing strategy, research performed on Big Red II simply cannot be done in cloud facilities. Cloud computing is great for scientific applications that take just a single computer processor, but cloud facilities do not have the sort of internal communication and data movement capabilities that Big Red II will have.
  • In “supercomputer years,” Big Red is 150 or so. Large supercomputers take months to build, install, and test. As they age, they tend to suffer hardware failures, which can be dramatic – even catastrophic – if a supercomputer is kept in operation too long. A dramatic Big Red failure, without a replacement on the way, could critically damage IU research progress and risk the ability of IU researchers to meet commitments to federal funding agencies.
  • We expect to operate Big Red II for five or six years, perhaps with a refresh of its internal processors during its lifetime. IU has excellent systems administration staff. By buying early and monitoring changing trends in technology, IU often stretches the useful life of a supercomputer more than our peer institutions.