NVIDIA: Introducing Tesla!
The products of the Tesla family were created to support the work of scientists and engineers.
The so-called General Purpose Graphics Processing Unit, or Tesla family of GPGPU products, also outperforms the already well-known NVIDIA Quadro devices in terms of performance. Thanks to their huge computing power, general-purpose graphics processing units not only have enough potential to run graphics applications, but are also the perfect solution for performing computational tasks in various disciplines.
The Tesla family currently consists of three members:
- C870: PCF Express x80-based expansion card based on GeForce G16 technology, with the exception of the plastic cover, with either the 8800 GTX or Quadro FX with 5600 could also be confused. The card, which receives power through two connectors, produces 170 watts of power at maximum load and boasts 518 gigaflops of computing power.
- D870: a With Quadro Plex the desktop, which is exactly the same on the outside, features two Tesla GPUs that can handle 1,5GB of GDDR3 graphics memory per head, with a combined computing power of over 1000 gigaflops.
- S870: the device, which can be inserted into a standard 19-inch 1U socket, already contains 4 Tesla GPUs, so the computing power and the size of the graphics memory are quadrupled accordingly, the average and maximum consumption values are around 550 and 800 watts, respectively.
NVIDIA is the CUDA for Tesla products introduced earlier this year (Complete Unified Device Architecture) recommends a development environment for programmers. The C870, D870, and S870 are available at suggested retail prices of $ 1499, $ 7500, and $ 12, respectively.
