NVIDIA Tesla P100
The NVIDIA Tesla P100 is a data center graphics card based on the Pascal architecture. Released on 2016-06-20 with a launch price of $10,000.00, it features 3584 CUDA cores and 16 GB of HBM2 memory.
General info about Tesla P100
| Architecture |
Pascal (GP100) |
| Market Segment |
Data Center |
| Release Date |
2016-06-20 |
| Launch Price |
$10,000.00 |
| Manufacturing Process |
16nm |
Technical specs of Tesla P100
| CUDA Cores |
3584 |
| Base Clock Speed |
1190 MHz |
| Boost Clock Speed |
1328 MHz |
| Transistor Count |
15.3B |
| VRAM Capacity |
16 GB HBM2 |
| Memory Bus Width |
4096 bits |
| Memory Bandwidth |
732.0 GB/s |
| TDP |
250 W |
Key technical parameters of the NVIDIA Tesla P100 include its 3584 CUDA cores, 1190 MHz base clock, and 1328 MHz boost clock, delivering high performance for data center applications.
Benchmark performance of Tesla P100
| FP32 (float) |
9.3 TFLOPS |
| FP16 (half) |
18.7 TFLOPS |
The NVIDIA Tesla P100 achieves 9.3 TFLOPS in FP32 performance, making it a top contender in data center workloads.
Gaming performance of Tesla P100
The NVIDIA Tesla P100 is not designed for gaming.
Features and connectivity of Tesla P100
| Tensor_cores |
none |
| Nvlink |
True |
| Ports |
PCIe Gen3 |
Physical dimensions of Tesla P100
| Length |
10.5in |
| Width |
4.4in |
| Height |
2-slot |
Rankings and efficiency of Tesla P100
| Ranking Position |
#5 |
| Popularity Ranking |
#10 |
| Cost Effectiveness |
1.2/5 |
| Power Efficiency |
1.6/5 |