 |
NVIDIA TESLA
|
| High Performance Computing - Supercomputing with NVIDIA Tesla GPUs |
| GPU acceleration is to use a graphics processing unit (GPU) together with a CPU to accelerate computational applications. GPU accelerators can provide unprecedented application performance by offloading compute-intensive portions of the workload to the GPU, while the remainder of the code still runs on the GPU. |
| NVIDIA Tesla GPU accelerators, based on the NVIDIA Kepler architecture, are designed to offer faster and more efficient compute performance for the most demanding computing applications in fields including: machine learning and data analytics, seismic processing, computational biology, chemistry, weather and climate modeling, signal processing, computational finance, physics, CAE and CFD. |
 |
|
The New NVIDIA Tesla GPU Accelerator- Tesla K80
|
| NVIDIA Tesla GPU accelerators, based on the NVIDIA Kepler architecture, are designed to offer faster and more efficient compute performance for the most demanding computing applications in fields including: machine learning and data analytics, seismic processing, computational biology, chemistry, weather and climate modeling, signal processing, computational finance, physics, CAE and CFD. |
 |
|
|
The Tesla K80 features:
• GPU Boost - Dynamically scales clocks for maximum application performance
• Zero-power Idle - Increase data center energy efficiency by powering down idle GPUs
• Double shared memory and register file - Increase 2x effective bandwidth and shared memory compared to the Tesla K20X and K10
• Multi-GPU Hyper-Q - Efficiently and easily schedule MPI ranks across GPUs |
 |
NVIDIA Tesla Server Solutions
|
| Tesla K80 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
2 Kepler GK210 |
480 GB/sec |
24 GB GDDR5 |
4992 |
 |
|
| Tesla K40 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Kepler GK110B |
288 GB/sec |
12 GB GDDR5 |
2880 |
|
|
| Tesla K20X |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Kepler GK110 |
250 GB/sec |
6 GB GDDR5 |
2688 |
|
|
| Tesla K20 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Kepler GK110 |
208 GB/sec |
5 GB GDDR5 |
2496 |
|
|
| Tesla K10 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
|
2 Kepler GK104s |
320 GB/sec |
8 GB GDDR5
(4 GB per GPU) |
3072
(1536 per GPU) |
|
|
NVIDIA Tesla Workstation Solutions
|
| NVIDIA® Tesla® GPU Accelerators turn standard PCs and workstations into personal supercomputers. Powered by CUDA® - the world’s most pervasive parallel-computing model Tesla GPU Accelerators for workstations deliver cluster level performance right at your desk.
GPU Computing Applications: Reservoir simulation, CAE (structural analysis), Molecular dynamics, Numerical analytics, Computational visualization (ray tracing)
|
| Tesla K40 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Kepler GK110B |
288 GB/sec |
12 GB GDDR5 |
2880 |
|
|
| Tesla K20 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Kepler GK110 |
208 GB/sec |
5 GB GDDR5 |
2496 |
|
|
| Tesla C2075 |
GPU Type |
Memory Bandwidth |
Memory Size |
CUDA Cores |
|
 |
1 Fermi GPU |
148 GB/sec |
6 GB GDDR5 |
448 |
|
|
| |
Email
Phone
