20 Petaflop Titan Supercomputer Set To Aid Scientific Research

World’s fastest open-science supercomputer uses CPU-GPU combination to improve energy effiency

The Titan supercomputer, the world’s fastest scientific computer, has gone live at the Oak Ridge National Laboratory (ORNL) in Tennessee, offering performance of 20 Petaflops, using GPUs (graphics processing units) alongside conventional processors.

Titan’s 20 petaflop top speed, means it can manage 20,000 trillion calculations every second. ORNL is part of the US Department of Energy’s network of research labs, so Titan will be open to researchers from academia, government laboratories and a number of other industries.

Potential uses for the Titan supercomputer include tackling climate change through the investigation of alternative energy sources, finding cures for disease and accelerating research in material science and astrophysics. The use of GPUs in supercomputers is becoming more widespread: the chips, originally designed to speed up image handling particulary for games machines, have evolved to more general-purpose uses.

Titan Supercomputer

“Science and technology have always been our primary goal, and Titan is a groundbreaking tool that will allow researchers worldwide to leverage GPU-accelerated computing to make unparalleled breakthroughs,” said Jeff Nichols, associate laboratory director for computing and computational sciences at ORNL.

Development began three years ago when ORNL decided to upgrade its 2.3 petaflop Jaguar system, which was at one point the world’s fastest supercomputer. Titan is ten times faster and five times more efficient than its predecessor, but it occupies the same amount of floor space and only slightly more energy.

The Cray XK7 system contains 18,688 nodes, with each holding a 16-core AMD Opteron 6274 processor and an Nvidia Tesla K20 graphics processing unit (GPU) accelerator. Titan also uses 710 terabytes of memory.

Energy efficiency

“One challenge in supercomputers today is power consumption,” added Nichols. “Combining GPUs and CPUs in a single system requires less power than CPUs alone and is a responsible move toward lowering our carbon footprint.  Titan will provide unprecedented computing power for research in energy, climate change, materials and other disciplines to enable scientific leadership.”

Nvidia claims that has ORNL simply upgraded Jaguar by expanding its CPU-based architecture, it would have been four times the current size and consume more than 30 megawatts of power.

“Basing Titan on Tesla GPUs allows ORNL to run phenomenally complex applications at scale, and validates the use of accelerated computing to address our most pressing scientific problems,” said Steve Scott, chief technology officer of the GPU Accelerated Computing business at Nvidia. “You simply can’t get these levels of performance, power- and cost-efficiency with conventional CPU-based architectures. Accelerated computing is the best and most realistic approach to enable exascale performance levels within the next decade.”

Do you know all about Green IT? Find out with our quiz!