|
April 2018 |
|
HPC data centers need to support the ever-growing demands of
scientists and researchers while staying within a tight budget. The
old approach of deploying lots of commodity compute nodes requires
huge interconnect overhead that substantially increases costs
without proportionally increasing performance.
NVIDIA Tesla P100 GPU accelerators are the most advanced ever
built, powered by the breakthrough NVIDIA Pascal™ architecture
and designed to boost throughput and save money for HPC and
hyperscale data centers. The newest addition to this family, Tesla
P100 for PCIe enables a single node to replace half a rack of
commodity CPU nodes by delivering lightning-fast performance in a
broad range of HPC applications.
TESLA P100 Accelerator Features and Benefits
PASCAL ARCHITECTURE
|
More than 18.7 TeraFLOPS of FP16, 4.7 TeraFLOPS of double-precision, and 9.3 TeraFLOPS of singleprecision performance powers new possibilities in deep learning and HPC workloads. |
| COWOS HBM2 |
Compute and data are integrated on the same package using Chip-on- Wafer-on-Substrate with HBM2 technology for 3X memory performance over the previous-generation architecture. |
| PAGE MIGRATION ENGINE |
Simpler programming and computing performance tuning means that applications can now scale beyond the GPU’s physical memory size to virtually limitless levels. |
Specifications:
| Specification |
Description |
GPU Architecture
|
NVIDIA Pascal™
|
Cuda Cores
|
3584 |
Single-Precision Performance
|
9.3 TeraFLOPS |
| Double-Precision Performance |
4.7 TeraFLOPS |
GPU Memory
|
16GB GDDR5
|
Memory bus width
|
4096-bit
|
Memory bandwidth
|
732 GB/s
|
Memory clock
|
715 MHz
|
System Interface
|
Full Height/Length PCI Express Gen3
|
Max Power
|
250W
|
Form Factor
|
111.15 mm, (4.38 inches) (H) × 266.70 mm, (10.5 inches) (L)
|
|
