NVIDIA Tesla
High Performance Computing (HPC) with NVIDIA Tesla GPU
WHY CHOOSE TESLA
DESIGNED FOR HPC PERFORMANCE AND RELIABILITY
The NVIDIA CUDA® parallel computing platform is enabled on GeForce®,Quadro®, and Tesla® products. Whereas GeForce and Quadro are designed for consumer graphics and professional visualization, respectively, the Tesla product family is designed ground-up for parallel computing and programming and offers exclusive High performance computing features.
HIGHEST PERFORMANCE FOR HPC APPLICATIONS
Tesla products are designed with exclusive features to maximize performance for supercomputing professionals:
Full double precision floating point performance
- 1.31 TFlops on the Tesla K20X
- Higher double precision than consumer products
Faster PCIe communication
- The only NVIDIA product with two DMA engines for bi-directional PCIe communication
Higher performance on technical applications with large data sets
- Larger on-board memory (6GB on a Tesla K20X and 8GB on a Tesla K10 GPU)
Faster communication with InfiniBand using NVIDIA GPUDirect™
- Special Linux patch, InfiniBand driver, and CUDA driver
BUILT FOR DATA RELIABILITY
Computational professionals depend on mission critical applications to run reliably without data error. All Tesla products undergo zero error tolerance testing in the most strenuous conditions for hundreds of times longer than any other NVIDIA products. Tesla exclusive features include:
ECC protection for uncompromised data reliability
- For memories inside the GPU and the external GDDR5 memory
Zero error tolerance stress testing
- Stress tested in manufacturing for several days
Manufactured by NVIDIA and guaranteed to be of highest quality
- Higher margin in memory and core clocks for added reliability
Enterprise level support
- Three year warranty, higher priority bug resolutions and feature requests, and product lifecycle of 18-24 months
ISV Certification
- HPC software vendors only certify software on Tesla products
BUILT FOR HPC SOLUTIONS
HPC customers have unique requirements. Only the Tesla business unit provides products and solutions to meet these special needs:
Integrated OEM Systems
- Innovative OEM servers, blades, and workstations designed specifically for Tesla products
Long-term supply and availability of Tesla products
- 18-24 month life-cycle availability
Data center warranty
- 3 year warranty from NVIDIA and OEM partners on Tesla products
Cluster management and GPU monitoring software
- GPU temperature monitoring, fan speed, and power
- Exclusive access to GPUs in a cluster
HPC expertise in pre-sales and post-sales support
- Total HPC solution expertise provided by CUDA/OpenCL tuning engineers, solution architects, and dedicated sales team