Desktop PC Supercomputer: Is It Real?
Is it possible to turn a personal computer into a powerful solution for specialized needs - a supercomputer? Yes, such an opportunity appeared, however, this is not exactly the path that analysts had predicted at the end of the last century. Over the past five to six years, video cards have gone from the main design modules - central processors significantly ahead. This led to the fact that they began to be used for unusual earlier needs.
The idea of calculating the GPU is not new. Last year, ATI, and now AMD has raised the idea of using graphics processors for specific calculations. Currently, in the role of computing devices, video cards are used to process enormous amounts of data obtained from geological surveys of the American oil company Hess. You can also call the project Folding @ Home, which is also an indicator of the use of GPU in the calculations, this time scientific. This project assumed the use of ATI chips, however, at the end of last year, Stanford researchers announced their interest in using nVidia GeForce 8800 chips.
To promote the idea of using GPUs as specialized computing solutions, nVidia has released the first public beta of the CUDA developers package. This SDK allows you to simplify the development of software that is able to use the video card family GeForce 8800 for any calculations. Recall that the GeForce 8800 GTX contains 128 "stream processors", which, according to nVidia, are capable of providing peak theoretical performance in the region of 520 GFlops, while using two video cards you can get almost 1 TFlops.
Even if we take into account the huge distance between theory and practice, the actual performance will still be at a very high level. It can be interesting for corporate clients, but the costs of building infrastructure can stun any large multinational corporation. It will take more than one year before the GPU will be applied by the corporate sector.
Source: www.thg.ru/technews/20070220_085307.html
The idea of calculating the GPU is not new. Last year, ATI, and now AMD has raised the idea of using graphics processors for specific calculations. Currently, in the role of computing devices, video cards are used to process enormous amounts of data obtained from geological surveys of the American oil company Hess. You can also call the project Folding @ Home, which is also an indicator of the use of GPU in the calculations, this time scientific. This project assumed the use of ATI chips, however, at the end of last year, Stanford researchers announced their interest in using nVidia GeForce 8800 chips.
To promote the idea of using GPUs as specialized computing solutions, nVidia has released the first public beta of the CUDA developers package. This SDK allows you to simplify the development of software that is able to use the video card family GeForce 8800 for any calculations. Recall that the GeForce 8800 GTX contains 128 "stream processors", which, according to nVidia, are capable of providing peak theoretical performance in the region of 520 GFlops, while using two video cards you can get almost 1 TFlops.
Even if we take into account the huge distance between theory and practice, the actual performance will still be at a very high level. It can be interesting for corporate clients, but the costs of building infrastructure can stun any large multinational corporation. It will take more than one year before the GPU will be applied by the corporate sector.
Source: www.thg.ru/technews/20070220_085307.html
')
Source: https://habr.com/ru/post/3468/
All Articles