The NVIDIA Tesla P4 is an enterprise grade GPU adapter that is designed for 24x7 operation, scalable up to six physical adapters in the R740 platform.
Our configuration of the R740xd has two available PCIe slots due to the high drive density, the BOSS PCIe drive adapter, and the 10G SPF+ network adapter.
The NVIDIA Tesla P4 is powered by the NVIDIA Pascal architecture and is purpose-built to boost efficiency for scale-out servers running deep learning workloads, enabling smart responsive AI-based services.
Features
The Nvidia adapter includes the following features:
- Small form-factor, 50/75-Watt design fits any scale-out server
- INT8 operations, which reduce latency by a multiple of 15
- A hardware-decode engine capable of transcoding and inferencing 35 HD video streams in real time
GPU architecture | NVIDIA Pascal |
Single-Precision Performance | 5.5 TeraFLOPS* |
Integer Operations (INT8) | 22 TOPS* (Tera- Operations per Second) |
GPU Memory | 8 GB |
Memory Bandwidth | 192 GB/s |
System Interface | Low-Profile PCI Express Form Factor |
Max Power | 50W/75W |
Enhanced Programmability with Page Migration Engine | Yes |
Server-Optimized for Data Center Deployment | Yes |
Hardware-Accelerated Video Engine | 1x Decode Engine, 2x Encode Engine |