Super Computers

SuperComputers Introduction:

A Computer is an electronic device that can perform various tasks quickly but a SuperComputer is an electronic device that can perform large number of tasks million times quicker than a normal Computer. As the world is evolving to a technology world the people are creating miraculous products that can be helpful in many ways. Likewise the SuperComputers which is a machine or can be said as a large Computer designed with most sophisticated microchips to solve the most complex scientific problems easily and quickly. When compare a Computer and a SuperComputer, the transistors, circuits, processors are all the same but what matters is the speed and memory size of a SuperComputer with numbers of components fixed inside it to work faster than ever.

A SuperComputer is a Computer machine designed to work millions times faster than a normal desktop Computer. It is used for various scientific and engineering applications that can handle large database and execute the programs faster. In the mid 1970’s the first super computer was developed by Cray Research which computes faster using a supercomputer form known as Vector processing, but initially Parallel processing is used which was developed in 1980 that can use multiple processors for faster execution of instructions. SuperComputers are very powerful and expensive too that can used by researchers for weather forecasting, scientific calculations, animated graphics, nuclear energy research, petroleum exploration and many more.

Serial and Parallel Processing

SuperComputers uses serial processing or vector processing at first in 1970’s, which can perform one work at a single period of time. Since it is time consuming and also a lengthy process which allows the researchers to do parallel processing. Parallel processing was introduced in 1980’s and since then it has been used by various SuperComputers that can work on many pieces at one time by splitting the pieces to many. Here the processors work in parallel to handle the works separately at same speed as on Serial processor.

Top Ten SuperComputers in the World:

There are many SuperComputers have been designed since the evolution of the first SuperComputer, CDC (Control Data Corporation) 6600 which was designed by Seymour Cray in 1964. It has single processor and priced at 8 million dollar at that time. But after than Microsoft’s CX1, IBM’s Roadrunner, Blue Gene and Watson’s have became popular in today’s world of SuperComputers. But here are some of the top ten SuperComputers in the World:-

• Sunway Taihu Light is a Chinese supercomputer developed by the National Research centre of parallel computer engineering and technology. It has a peak speed of 93.01 PFLOPS means it can make 93 quadrillions calculations per second. It is designed completely using processors and has the highest processing capacity of 125.436 PFLOPS per second. The supercomputer can be used in various scientific research, manufacturing, data analytics, weather forecasting and also other engineering works.

• Tianhe-2 is another supercomputer from China developed by National university of Defence technology with a team of 1300 scientists and engineers. Its peak speed is 33.86 pflops or quadrillions of calculations per second. Tianhe-2 is designed to run Intel Xeon E5 processors, custom processors and Intel Xeon Phi processors. It has more than 30 lacs cores which tends to perform faster than any other supercomputers. The memory storage capacity is 1,024,000GB and runs on Kylin Linux Operating system.

• Cray Titan is a supercomputer built by Cray at Oak Ridge National laboratory, United States to be used in various science projects. It has 16-core AMD Opteron CPU and Nvidia GPUs to perform much faster while developing for gaming systems. Its peak speed is 17.59 PFLOPS or more than 17,000 trillion calculations per second. The supercomputer features 18,688 compute nodes, 710 TB memory and 561,000 cores that makes it a powerful and superfast computers.

• IBM Sequoia is a supercomputer from IBM lists that has 1.6 million cores combines and uses 1,572,864 cores to give 17.17 PFLOPS performance. It is based on IBM Bluegene framework which is another supercomputer used earlier. It is very fast and can run 17,000 trillions of calculations per second. This is also used by researchers and scientists for various astronomy fields, climate and nuclear energy.

• Fujitsu K Computer is the only supercomputer designed by Fujitsu at RIKEN Advanced Institute for Computational sciences in Kobe Japan. It is listed as the top ten powerful computers in the world due to its peak speed of 10.51FLOPS and uses 705,024 processing cores. The machine runs on Fujitsu’s own SPARC64 an octa-core processor that gives the best performance.

• Mira is an IBM’s Blue Gene/Q line of supercomputers that has a peak speed of 10 PFLOPS or can perform 10 quadrillions per second. It has almost 800,000 processors and cores that will finish a job in seconds. It is mainly designed for the scientists to help them study exploding stars, nuclear energy, weather forecast, jet engines and lots more.

• Piz Daint is Europe’s most powerful supercomputer with a peak speed of almost 7 PFLOPS and uses 73,808 Nvidia accelerated cores for best performance. Based on Cray XC30 supercomputer, Piz Daint incorporates Cray’s data warp technology for long term storage. It will provide the users two computing nodes with hybrid CPU-GPU and CPU only nodes for researching intensive CPU simulations, material science, geophysics, life sciences and many more.

• Shaheen II is a supercomputer from Saudi Arabia which is also based on Cray XC30 that can deliver 5.536PFLOPS of sustained LINPACK performance. It has 6174 dual sockets compute nodes based on 16 cores Intel Haswell processors running at 2.3GHz speed. With 197,568 processor cores and 790 TB memory is more powerful supercomputer in this region.

• Stampede is a Dell Poweredge C8220 system based around powerful interlinked desktop Computers with 5.17PFLOPS peak performance. It has 6400 nodes with high performance processors, RAM and memory storage for easy scientific computations. It has 522080 cores that can execute code orders magnitude faster and 270 TB total memory storage capacity.

• Juqueen is a Europe’s most powerful supercomputer and is also listed in top ten with a peak performance of 5.01PFLOPS. It is based on IBM’s Blue Gene architecture with 28,672 nodes and 458,752 processing cores. It uses IBM powerPC A2 computer card with 1.6GHz processor speed, 16 cores per node, 16GB RAM per node and all total memory storage capacity of 448 TB.