NVIDIA GPUs power world's fastest supercomputer

The supercomputer was built by the National University of Defense Technology and is located at the National Supercomputing Center in Tianjin, China. According toNVIDIA, the computer is 30 percent faster than the world’s second largest computer which is at the Oak Ridge National Laboratory in Tennessee.

Tianhe-1A supercomputer parallels large quantities of GPUs with multi-core CPUs to significantly boost performance, power and size. The Tianhe-1A uses 7,168 NVIDIA Tesla M2050 GPUs and 14,336 CPUs. This would be equivalent to more than 50,000 CPUs and would require twice as much floor space to deliver the same performance using CPUs alone.

A 2.5 petaflop system using CPU’s only would require more than 12 megawatts of power to run it. By using NVIDIA’s GPUs in a heterogeneous computing environment, Tianhe-1A consumes only 4.04 megawatts, making it 3 times more power efficient.

At the HPC 2010 in China, Guangming Liu, chief of NationalSupercomputerCenter in Tianjin stated:“The performance and efficiency of Tianhe-1A was simply not possible without GPUs. The scientific research that is now possible with a system of this scale is almost without limits; we could not be more pleased with the results."

Jen-Hsun Huang, president and CEO of NVIDIA commented,“GPUs are redefining high performance computing. With the Tianhe-1A, GPUs now power two of the top three fastest computers in the world today. TheseGPUsupercomputers are essential tools for scientists looking to turbo-charge their rate of discovery."

NVIDIA first invented the computer graphics chip in 1999 and showed the computer industry the power of computer graphics. NVIDIA’s programmable GPUs have made advancements in parallel processing which makes supercomputing inexpensive and widely accessible; NVIDIA holds more than 1,600 patents worldwide.

Source

IBM to ship z196 - world's fastest microprocessor

This world record-breaking speed is necessary for businesses managing huge workloads, such as banks and retailers, especially as the world becomes increasingly more inter-connected, data has grown beyond the world'sstorage capacity, and business transactions continue to skyrocket.

For example, according to a study by Berg Insight, the number of active users of mobile banking and related financial services worldwide is forecasted to increase from 55 million in 2009 to 894 million in 2015.

Such trends are driving the need for innovation in systems that can help clients take advantage of them to provide new services and develop new business models. In response, IBM continues to invest in systems innovation -- delivering workload-optimized systems with innovation in microprocessors, software and hardware.

The new zEnterprise technology is the result of an investment of more than $1.5 billion in IBM research and development in the zEnterprise line, as well as more than three years of collaboration with some of IBM's top clients around the world.

You need Flash installed to watch this video

z196 Processor

The z196 processor is a four-core chip that contains 1.4 billion transistors on a 512-square millimeter (mm) surface. The chip was designed by IBM engineers in Poughkeepsie, N.Y., and was manufactured using IBM's 45 nanometer (nm) SOI processor technology in the company's 300mm fab in East Fishkill, N.Y. There were also major contributions to the z196 processor development from IBM labs in Austin, TX, Germany, Israel and India.

The mainframe processor makes use of IBM's patented embedded DRAM (eDRAM) technology, which allows IBM to place denseDRAMcaches, or components, on the same chips as high-speed microprocessors, resulting in improved performance.

The New zEnterprise 196

From a performance standpoint, the zEnterprise System is the most powerful commercial IBM system ever. The core server in the zEnterprise System -- called zEnterprise 196 -- contains 96 of the world's fastest, most powerful microprocessors, capable of executing more than 50 billion instructions per second. That's roughly 17,000 times more instructions than the Model 91, the high-end of IBM's popular System/360 family, could execute in 1970.

This new IBM microprocessor technology has new software to optimize performance of data-heavy workloads, including up to a 60% improvement in data intensive and Java workloads. Increased levels of system performance, in turn, increases software performance, which can reduce software license costs.

The new system offers 60% more capacity than its predecessor, the System z10, and uses about the same amount of electricity.

Energy efficiencies were achieved through advances inmicroprocessordesign, 45nm silicon technology, more efficient power conversion and distribution, as well as advanced sensors and cooling control firmware that monitors and makes adjustments based on environmental factors such as temperature and humidity levels and even air density.

The combination of zEnterprise chip speed, memory, system reliability, availability, security and storage architecture provide an optimal environment for managing the world's most demanding workloads.

Source