Top 10 Supercomputers

The SuperMUC performs at 2.9 petaflops, thanks to about 150,000 processing cores. Despite this incredible processing power, it has been pushed from No. 4 on the June 2012 list all the way to No. 10. It's one of several IBM systems to make the list, but the SuperMUC is unique in a couple of different ways. Located in Germany's Leibniz Supercomputing Centre, the SuperMUC uses a new hot-water cooling system to keep the computer's brain from frying while it's performing billions upon billions of operations.

Efficiency is what really sets the SuperMUC apart: IBM says it's 40 percent more energy efficient than an air-cooled system would be. They claim the water removes heat 4,000 times more efficiently than air. Thanks to its cutting-edge hardware, the SuperMUC is Germany's second fastest supercomputer. In fact, it's the third fastest supercomputer in Europe, and nearly twice as fast as the previous year's No. 10.

Vulcan is the first of four computers in the top 10 running on IBM's powerful BlueGene/Q platform with Power BQC 16-core 1.6GHz processors. Besting the previous entry by over a petaflop, this powerhouse performs at 4.3 petaflops with its nearly 400,000 cores.

This supercomputer is one of two in the top 10 operating at the U.S. Department of Energy's National Nuclear Security Administration (NNSA) at the Livermore National Laboratory (LLNL) in California. In mid-2013, the lab opened up computer time on Vulcan to U.S. businesses for collaborative work on projects that will increase scientific and technological advancement, improve U.S. competitiveness and enhance the high-performance computing (HPC) workforce, all in exchange for a share of the operational costs. Vulcan is also intended for collaboration with academic and research institutions in areas such as energy, security, atmospheric science and bioscience.

JuQUEEN also runs on the IBM BlueGene/Q platform, with max performance of 5 petaflops on about 459,000 cores. While many computers on the list have been around for the last couple years, JuQUEEN was built in 2012 as a replacement for another system, JUGENE, which was the ninth fastest system on the November 2010 TOP500.

Researchers at the Jülich-Aachen Research Alliance can submit proposals for projects to justify using some of JuQUEEN's cores, and if approved, they get to carve out some time with one of the most powerful supercomputers in the world. Time on this and two other powerful German supercomputers, SuperMUC and HERMIT (39th on the TOP500 list), can be requested through the Gauss Centre for Supercomputing (GCS) and the Partnership for Advance Computing in Europe (PRACE).

Stampede made its way into the No. 7 spot in November 2012, rose to No. 6 in June 2013 and then resumed its former place on the most recent list. Funding for the system came from an award from the National Science Foundation (NSF), and Stampede is housed at the Texas Advanced Computing Center (TACC) at the University of Texas in Austin. This Dell PowerEdge system runs Intel Xeon E5 8-core 2.7GHz processors in conjunction with Intel Xeon Phi coprocessors. Its roughly 462,000 cores allow it to achieve nearly 5.2 petaflops in performance.

Since January 2013, Stampede has been open for use by scientists and other researchers in all fields as part of the NSF's Extreme Science and Engineering Discovery Environment (XSEDE) program, a virtual system for openly sharing the computing power of 16 supercomputers and related resources. Ninety percent of the system is dedicated to XSEDE and the rest is at the discretion of the TACC's director. Qualified researchers at any U.S. institution can submit proposals via the XSEDE Web site.

Although Piz Daint has been in operation since April 2013, this Cray system later went through a major upgrade that boosted it into the No. 6 position and dethroned JuQUEEN as the most powerful supercomputer in Europe.

Piz Daint runs Intel Xeon E5 processors along with NVIDIA graphical processing units (GPUs) for added performance, allowing it to reach 6.3 petaflops with its 116,000 processing cores. It resides at the Swiss National Supercomputing Centre (CSCS) and will be put to use modeling weather and climate patterns and performing scientific computation in a variety of other fields.

The Piz Daint is also one of the most energy efficient supercomputers on the list, with an efficiency level of 3,185.9 megaflops per watt (MFLOPS/W) [sources: Green500, Smith]. It's the only supercomputer to make the top 10 in both the TOP500 and GREEN500 lists. GREEN500 takes all the supercomputers in the TOP500 list and ranks them by energy efficiency. Piz Daint's hybrid architecture, which uses both traditional CPUs (central processing units) and more energy-efficient GPUs, helps keep its energy usage low.

IBM's Mira, which became fully operational in 2013, peaks at a performance of 8.6 petaflops. That's more than 2 petaflops over the Piz Daint, and nearly 3.5 petaflops more than Stampede.

Mira runs on 786,000 processor cores. It's located at the Argonne National Laboratory, a research laboratory for the U.S. Department of Energy. It uses IBM's BlueGene/Q platform and replaces an older IBM system, Intrepid, which ranked fourth on the list in 2008.

Researchers who submit proposals for the Innovative and Novel Computational Impact on Theory and Experiment program (INCITE) through the U.S. Department of Energy's Office of Science can claim processor time on Mira. Sixty percent of the computer's capacity goes to their research, while 30 percent goes towards the Advanced Science Computing Research Leadership Computing Challenge. The final 10 percent is reserved for urgent, time-sensitive computations [source: Information Week].

Fujitsu's K computer, the only supercomputer in Japan that made the top 10, reigned as the fastest supercomputer in the world on both 2011 lists, but has since been edged down to No. 4. Still, it breaks the single-digit barrier and makes a noticeable jump in speed over IBM's Mira with a performance of 10.5 petaflops.

The K computer is located at Japan's RIKEN Advanced Institute for Computational Science, where it performs scientific operations including global disaster prevention, meteorology and medical research [source: Fujitsu]. Unlike many of the other supercomputers on the list, it doesn't run on IBM architecture. The K computer uses Fujitsu's own SPARC64 VIIIfx octo-core processors. Its 705,000 computer cores help it churn through operations at an incredible pace.

But believe it or not, the last three fastest supercomputers are leaps and bounds more powerful than the K computer.

Sequoia was the top ranked supercomputer on the June 2012 TOP500 list, but dropped to No. 2 in November 2012, and now rests at No. 3. It's still no slouch, thanks to around 1.6 million processing cores that can crank out an incredible 17.2 petaflops of performance. Wondering just how incredible that is?

Well, if we look back less than a decade ago to 2008, IBM's Roadrunner made history (and grabbed the top slot) for cracking 1 petaflop, aka performing 1,000 trillion operations per second [source: IBM]. IBM said Roadrunner was equivalent to 100,000 of 2008's laptops in performance. And Sequoia is 17 times as fast! Sequoia is one of four computers on the November 2013 list running on the BlueGene/Q IBM design, a 16-core 1.6GHz chip. That's not an especially fast clock speed by today's standards, but with 96 racks of chips, the performance really adds up.

What's Sequoia doing with all that speed, anyway? Sequoia is 63 percent faster than the fourth fastest computer on the list, and IBM is putting Sequoia to work, of course. Like Vulcan, it operates at the U.S. Department of Energy's National Nuclear Security Administration at the Livermore National Laboratory. The computer's doing important (and classified) work: One of its responsibilities is simulating nuclear explosions.

One of the two Cray systems on the List, the Titan lives up to its name, utilizing Opteron 6274 16-core 2.2GHz processors along with NVIDIA GPUs to perform at an amazing 17.6 petaflops on around 561,000 cores. It's located at the Oak Ridge National Laboratory (ORNL) run by the Oak Ridge Leadership Computing Facility (OLCF). Titan nabbed the No. 1 spot on the Top500 the month after it was deployed in late 2012, but dropped down to No. 2 in the June 2013 rankings, where it remains.

Titan is technically a major upgrade to the ORNL's previous supercomputer, Jaguar. Due to its novel hybrid architecture, Titan fits into the same cabinets as Jaguar while achieving nine times Jaguar's speed with only around a 60 percent increase in energy consumption.

Titan is one of two supercomputers in the top 10 that incorporates NVIDIA GPUs to increase performance while keeping the power consumption lower than it would be for a similarly powerful all-CPU system. Utilizing the GPUs to run applications at the speed of which the system is capable requires a new approach to programming its software, however. To address this, the OLCF has partnered with Cray and NVIDIA to create the Center for Accelerated Application Readiness (CAAR), which is hard at work coming up with best practices for code development.

As with Mira, researchers can submit proposals through the U.S. Department of Energy's INCITE program to earn time on this workhorse.

The Tianhe-2 (which means Milky Way-2) appeared on the scene in 2012, more than two years ahead of schedule, and leapt into the No. 1 spot. It processes at an astounding 33.9 petaflops, nearly twice the performance of the Titan or Sequoia, and more than 10 times the performance of Tianhe-1A, which held the No. 10 spot in June 2013.

The system runs on a mix of Intel Xeon E5 processors, custom processors and Intel Xeon Phi coprocessors: approximately 3,120,000 cores in total.

Developed by China's National University of Defense Technology (NUDT) and located at the National Super Computer Center in Guangzhou, the system will be used for education and research. It is currently China's only top-10 entry, but the Tianhe-2 gives them bragging rights for sheer processing power in a single machine.

Tianhe-2 runs on a custom version of the Ubuntu Linux operating system called Kylin, which was developed through a partnership between the NUDT, the China Software and Integrated Circuit Promotions Centre (CSIP) and Canonical (creators of Ubuntu). In fact, all of the top 10 supercomputers, and most of the top 500, run some flavor of Linux.

Kylin isn't just for supercomputers. It is a freely available, open-source operating system tailored specifically for Chinese users, and can be downloaded from Ubuntu's site for use on personal computers.