Two supercomputing center was designed by engineers in the United States designed capable of doing 20 petaflop (20 trillion calculations / calculations per second). Speed would be far above the Tianhe-1, the name of China's supercomputer that perched on the top 500 list of fastest computers today with a performance of up to 2.57 petaflop.
One of the supercomputer built at Oak Ridge National Laboratory, the same location as the Jaguar supercomputer which ranked as two of the world today. The second supercomputer built at Lawrence Livermore National Laboratory by IBM. However, both systems are estimated to be completed in 2012.
In the Top 500 list has just been released, China is not only recorded the fastest supercomputers, but also include 42 supercomputers in it beyond the developed countries like Japan, Germany, and England. America still dominates with 275 of the 500 supercomputers on the list.
The new Intel processor is still the choice of most supercomputers. From 500, 398 using a series of Intel processors. While AMD's only used 57 supercomputers. Most of the included Chinese-made and still use the chip American companies, including such artificial support Nvidia GPU.KOMPAS.com.
 | ASUS P6T7 WS SuperComputer - Motherboard - SSI CEB - iX58 - LGA1366 Socket - SATA-300 (RAID), SAS, eSATA - 2 x Gigabit Ethernet - FireWire - HD Audio (8-channel) ASUS P6T7 WS SuperComputer The ASUS P6T7 WS SuperComputer is the ideal foundation for a powerful PC. It delivers awesome power, dependable performance and unparallel multiple I/O scalability for the most demanding tasks and future upgrades. See detail here..  |
 | Computer Architecture: From Microprocessors to Supercomputers (Oxford Series in Electrical and Computer Engineering) Computer Architecture: From Microprocessors to Supercomputers provides a comprehensive introduction to computer architecture, covering topics from the design of simple microprocessors to techniques used in the most advanced supercomputers. See detail here..  |
 | Understanding Supercomputing Issues include how to cluster computers into a functional unit, that can run these all efficiently (load balance). Or, how to make novel arrangements of computers into a supercomputer. See detail here..  |