Visit additional Tabor Communication Publications
November 16, 2011
WEST LAFAYETTE, Ind., Nov. 15 -- The nation's fastest campus supercomputer has been built at Purdue University.
Nicknamed "Carter," the supercomputer was ranked 54th this week on the latest international Top500.org list of the world's most powerful supercomputers.
The supercomputer was built using the latest technologies from Intel, HP and Mellanox, including not-yet-released Xenon E-5 "Sandy Bridge" Intel processors and HP Proliant servers.
The supercomputer was funded through a cooperative program in which faculty members pool research funds to purchase computing resources. Carter is the fastest supercomputer installed at a college campus that is not a part of a national center and is for campus use.
At Purdue, Carter is already being used for a variety of scientific research projects.
Alan Qi, an assistant professor in computer science, statistics and biology, is using Carter to identify cancer stem cells based on massive high-dimensional flow cytometry data. New theories about cancer therapies indicate that cancer growth can be more effectively curtailed by specific treatments targeted at cancer stem cells, which are responsible for growing new cancer cells.
"Currently the massive patient data is examined manually by a technician; this procedure is obviously very slow and prone to errors. Carter allows us to analyze the data by new advanced statistical models and to have more accurate results in a couple of hours," Qi said.
Michael Baldwin, an assistant professor of atmospheric science, is developing computational techniques to predict hazardous weather in the United States, such as tornadoes, hurricanes and blizzards. He is also developing improved modeling tools to determine how climate change affects severe weather.
"We need as much computing cabability as we can get to make our forecasts as detailed as possible," Baldwin says. "In our models, we combine physical processes of the atmosphere with high-resolution data from weather radar and satellites. We can use Carter to create useful forecasts that the public can use and potentially save lives.
"Carter is running twice as fast as the supercomputer we were using and is using only half of the nodes. That will allow us to scale our models for better forecasts."
In addition to these projects, Carter is also being used to study the quantum effects of future computer chips and to study the effects of greenhouse gases other than CO2 in the atmosphere.
"This groundbreaking supercomputer shows what is possible when researchers, campus IT staff and corporate partners work together as a close team," said Gerry McCartney, chief information officer, vice president for information technology at Purdue and the Olga Oesterle England Professor of Information Technology. "Carter is the fastest research computer at any institution that was completely paid for with faculty and university funds. It was built because Purdue, Intel, HP and Mellanox shared a vision to create a new resource for scientific discovery."
Constructing Carter was a team and global effort. The team that designed and built Carter was composed of six Purdue staff members; 11 HP staff members, including one person who contributed from Ireland; three Intel staff members, two of whom are based in Russia; and five staff members from Mellanox, four of whom traveled from Israel to work on site at Purdue.
Carter is named for Dennis Carter, a Purdue alumnus who earned his master's degree in electrical engineering in 1974. Carter was responsible for developing the innovative "Intel Inside" marketing campaign, which debuted in 1991.
Carter was built with a high-performance HP ProLiant server cluster based on the HP ProLiant SL6500s Scalable System with 648 server nodes featuring 1,296 future Intel Xeon E5 family processors totaling 10,368 cores. With these processors, Carter's servers can increase application performance up to 70 percent compared to the current generation.
"Purdue required a server platform that offered unmatched compute power without increasing system footprint or energy use," said Paul Santeler, vice president and general manager of Hyperscale Computing, HP Industry Standard Servers and Software. "We're delighted that HP ProLiant technology will deliver the exceptional performance needed to help accelerate Purdue's critical research initiatives in cancer treatment, meteorology and environmental sustainability."
At 186.9 teraflops capacity, this is the fifth Purdue supercomputer to be internationally ranked in the last four years, and Purdue has three supercomputers - Carter, Rossmann and Coates - on this year's rankings.
The TOP500 Supercomputer Sites project has been ranking the 500 most powerful known systems twice a year since 1993 as a way of detecting and tracking trends in high-performance computing. Rossmann and Coates placed 208 and 474 on the latest list.
Raj Hazra, general manager of high-performance computing for Intel, said Carter provides three times more performance than Purdue's 2008 supercomputer, Steele, while consuming less than half the energy and occupying half the space.
"These benefits are achieved thanks to numerous technology improvements that include Advanced Vector Extensions (AVX). The improvements found in our new microprocessors make them up to two times faster than our previous Xeon generation and much more power efficient," Hazra said.
"We at Intel are very excited as this is one of the very first supercomputers built on our upcoming Intel Xeon E5-2600 series processors. Being part of the Carter supercomputer project is an honor; learning it will be used to support numerous groundbreaking scientific projects is how we envision our technology helping to change the world."
In addition to the high-speed processors, the 648 nodes in Carter are connected via a Mellanox FDR InfiniBand network, which transfers data at 56 gigabits per second (GB/s). FDR InfiniBand doubles the throughput of data over previous designs, allowing Carter to process more data than was previously possible, said David Barzilai, vice president of marketing at Mellanox Technologies.
"We applaud Purdue's initiative to build the Carter supercomputer to drive the next era of scientific and medical research that exploit the performance and efficiency benefits of Mellanox's FDR 56Gb/s InfiniBand server and storage I/O solution," Barzilai said. "Mellanox's FDR InfiniBand I/O solution, with PCI-Express 3.0 support, delivers more than 12GB/s bi-directional throughput, two times higher compared to previous InfiniBand generations, reduces network latency to sub 1usec, and delivers application acceleration and offloading to achieve the highest scalability and system utilization. We look forward to continuing our collaborations with the university, Intel and HP to deliver exceptional performance for Purdue's research activities."
Purdue's "starting five of supercomputing" -- Carter, Hansen, Rossmann, Coates and Steele -- are all the result of Purdue's Community Cluster Program, which was the recipient of Campus Technology magazine's Innovators Award in 2010.
Source: Steve Tally, Purdue University
In a recent solicitation, the NSF laid out needs for furthering its scientific and engineering infrastructure with new tools to go beyond top performance, Having already delivered systems like Stampede and Blue Waters, they're turning an eye to solving data-intensive challenges. We spoke with the agency's Irene Qualters and Barry Schneider about..
Large-scale, worldwide scientific initiatives rely on some cloud-based system to both coordinate efforts and manage computational efforts at peak times that cannot be contained within the combined in-house HPC resources. Last week at Google I/O, Brookhaven National Lab’s Sergey Panitkin discussed the role of the Google Compute Engine in providing computational support to ATLAS, a detector of high-energy particles at the Large Hadron Collider (LHC).
The Xeon Phi coprocessor might be the new kid on the high performance block, but out of all first-rate kickers of the Intel tires, the Texas Advanced Computing Center (TACC) got the first real jab with its new top ten Stampede system.We talk with the center's Karl Schultz about the challenges of programming for Phi--but more specifically, the optimization...
May 23, 2013 |
he study of climate change is one of those scientific problems where it is almost essential to model the entire Earth to attain accurate results and make worthwhile predictions. In an attempt to make climate science more accessible to smaller research facilities, NASA introduced what they call ‘Climate in a Box,’ a system they note acts as a desktop supercomputer.
May 22, 2013 |
At some point in the not-too-distant future, building powerful, miniature computing systems will be considered a hobby for high schoolers, just as robotics or even Lego-building are today. That could be made possible through recent advancements made with the Raspberry Pi computers.
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.