NVIDIA Tesla P100 16GB CoWoS HBM2 PCIe 3.0 - Passive Cooling, GPU-NVTP100-16
BUS: PCI-E 3.0 16x Memory size: 16 GB Stream processors: 3584 Theoretical performance: TFLOP
All NVIDIA Ampere, NVIDIA Grace Hopper, and NVIDIA Ada Lovelace GPU architectures are subject to a non-cancellable, non-returnable (NCNR) period of 52 weeks. Additionally, the product is subject to sanctions for certain countries and the end customer must be documented.
We can supply these GPU cards directly and with an individual B2B price. Contact us with your inquiry today.
| Product code | 214.151623 |
|---|---|
| Part number | GPU-NVTP100-16 |
| EAN | 672042242443 |
| Manufacturer | NVIDIA |
| Availability | Not in stock |
| Warranty | 24 months |
| Weight | 1 kg |
| The price includes all legal fees | |
Detailed information
NVIDIA Tesla
Tesla products from the NVIDIA company introduce a line of computational graphics processors that are very similar to the NVIDIA Quadro series (they usually use the same chip). However, they have an isolated display interface. They're also available in passively-cooled form-factors, which are specifically appropriate for use in servers (rack-mounts).
CUDA Technology
Users of professional applications can, thanks to the CUDA architecture, use graphical CUDA stream processors. Thanks to this, it is possible to use the raw performance of a graphics card for specific calculations, which can significantly increase work speed compared to the use of a traditional processor, which are significantly limited by the lower number of cores.
Product Specifications
Graphical chip: Tesla P100
BUS: PCI Express x16 3.0
Memory size: 16GB
Parameters
| Product line | Tesla |
|---|---|
| Number of stream processors | 3584 |
| Power consumption (W) | 250 |
| Architecture | Pascal |
| Monitor output | None |
| Interface | PCI-E 3.0 16x |
| Gigabytes of memory | 16 |
| Cooling type | Passive |
| Memory type | HBM2 |
| Slot count | 2 |
| Profile | FH |
