Description
The A100 PCIe 80 GB is a professional graphics card by NVIDIA, launched on June 28th, 2021. Built on the 7 nm process, and based on the GA100 graphics processor, the card does not support DirectX. Since A100 PCIe 80 GB does not support DirectX 11 or DirectX 12, it might not be able to run all the latest games. The GA100 graphics processor is a large chip with a die area of 826 mm² and 54,200 million transistors. It features 6912 shading units, 432 texture mapping units, and 160 ROPs. Also included are 432 tensor cores which help improve the speed of machine learning applications. NVIDIA has paired 80 GB HBM2e memory with the A100 PCIe 80 GB, which are connected using a 5120-bit memory interface. The GPU is operating at a frequency of 1065 MHz, which can be boosted up to 1410 MHz, memory is running at 1512 MHz.
Being a dual-slot card, the NVIDIA A100 PCIe 80 GB draws power from an 8-pin EPS power connector, with power draw rated at 300 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. A100 PCIe 80 GB is connected to the rest of the system using a PCI-Express 4.0 x16 interface. The card measures 267 mm in length, 111 mm in width, and features a dual-slot cooling solution.
This new instance offers the following advantages over the previous generation:
- Up to three-times higher AI training performance for large language models (LLMs), deep learning recommendation models (DLRMs), and computer vision (CV) workloads
- Up to 1.25-times higher AI inference performance for workloads, such as real-time conversational AI, manufacturing inspection, and autonomous driving
- Up to 1.8-time higher performance on scientific HPC applications
Features
- NVIDIA Ampere GPU architecture
- Compute-optimized GPU
- 6912 / 3584 NVIDIA® CUDA® Cores
- 432 / 224 NVIDIA® Tensor Cores
- 80GB HBM2e / 24GB HBM2 memory with ECC
- Up to 1,935 / 933 GB/s memory bandwidth
Max. power consumption: 300W / 165W - Graphics bus: PCI-E 4.0 x16
- Thermal solution: Passive
Specifications
| General Information | |
|---|---|
| Manufacturer | NVIDIA Corporation |
| Manufacturer Part Number | 900-21001-0020-000 |
| Manufacturer Website Address | http://www.nvidia.com |
| Predecessor | Tesla Turing |
| Successor | Tesla Ada |
| Production | Active |
| Bus Interface | PCIe 4.0 x16 |
| Product Type | Graphic Card |
| Graphics Processor | |
| Chipset Manufacturer | NVIDIA |
| GPU Name | GA100 |
| Architecture | Ampere |
| Foundry | TSMC |
| Process Size | 7 nm |
| Transistors | 54,200 million |
| Density | 65.6M / mm² |
| Die Size | 826 mm² |
| Chip Package | BGA-2743 |
| Clock Speeds | |
| Base Clock | 1065 MHz |
| Boost Clock | 1410 MHz |
| Memory Clock | 1512 MHz 3 Gbps effective |
| Memory | |
| Memory Size | 80 GB |
| Memory Type | HBM2e |
| Memory Bus | 5120 bit |
| Bandwidth | 1,935 GB/s |
| Board Design | |
| Slot Width | Dual-slot |
| Length | 267 mm 10.5 inches |
| Width | 111 mm 4.4 inches |
| TDP | 300 W |
| Suggested PSU | 700 W |
| Outputs | No outputs |
| Power Connectors | 8-pin EPS |
| Board Number | P1001 SKU 200 |
| Render Config | |
| Shading Units | 6912 |
| TMUs | 432 |
| ROPs | 160 |
| SM Count | 108 |
| Tensor Cores | 432 |
| L1 Cache | 192 KB (per SM) |
| L2 Cache | 80 MB |
| Graphics Features | |
| DirectX | N/A |
| OpenGL | N/A |
| OpenCL | 3.0 |
| Vulkan | N/A |
| CUDA | 8.0 |
| Shader Model | N/A |
| Form Factor | Plug-in Card |
| Card Length | Full-length |
| Card Height | Full-height |
| Cooler Type | Passive Cooler |
| Theoretical Performance | |
| Pixel Rate | 225.6 GPixel/s |
| Texture Rate | 609.1 GTexel/s |
| BF16 | 311.84 TFLOPS (16:1) |
| TF32 | 155.92 TFLOPs (8:1) |
| FP64 Tensor | 19.49 TFLOPS (1:1) |
| FP16 (half) | 77.97 TFLOPS (4:1) |
| FP32 (float) | 19.49 TFLOPS |
| FP64 (double) | 9.746 TFLOPS(1:2) |
Reviews
There are no reviews yet.