Nvidia A100 Graphics Card
KWD 7,030KWD 8,329 (-16%)
- Architecture: NVIDIA Ampere
- Memory: 80GB HBM2e
- CUDA Cores: 6,912
- Tensor Cores: 640
- Performance: 312 TFLOPS (FP16)
- Interface: PCIe Gen4
- Power: 400W (SXM4) / 250W (PCIe)
- Applications: AI, deep learning, HPC
Description
Exploring GPUs for your elastic data centers in Kuwait? Do you want to buy graphics cards from authorized NVIDIA and AMD GPU store? Shop powerful GPUs for top-edge data centers at ServerBasket without worrying about investment and time to value secured by a 1-year warranty with dedicated 24/7 available technical experts. Choose the NVIDIA A100 80GB GPU built to accelerate the most advanced data centers with support for a wide set of maths precisions – FP32, FP64, FP16, and INT8 – guaranteeing an accelerator for every workload. The NVIDIA A100 Tensor Core PCIe GPU, with GDDR6 memory, is a proven engine for the most important requirements of our time including data analytics, IoT, AI, data simulation, HPC, and so on. The dual-slot 80GB GPU, based on the Ampere GA100 GPU, employs a passive cooling mechanism for maximized overclocking. It has a TDP of 300W to effortlessly accommodate diverse applications with dynamic computational speed and data throughput needs. Its 1.94TB/s memory bandwidth, the world’s highest for a PCIe GPU, enables lightning-fast time to solution for massive models and large data sets. Dynamically adjust to changing workload demands with MIG support on the A100 which allows you to create as many as seven isolated partitions virtually. Shop the NVIDIA A100 Tensor Core GPU in Kuwait now at the most competitive price.
Technical Specifications
Brand | NVIDIA |
---|---|
Product Name | NVIDIA A100 Tensor Core |
GPU architecture | NVIDIA Ampere |
GPU memory | 80GB |
CUDA Cores | 6912 |
Streaming Multiprocessors | 108 |
Tensor Cores | Gen 3 | 432 |
FP64 | 9.7 TFLOPS |
FP64 Tensor Core | 19.5 TFLOPS |
FP32 | 19.5 TFLOPS |
GPU Memory | 80GB HBM2e |
GPU Memory Bandwidth | 1,935 GB/s |
Max Thermal Design Power (TDP) | 300W |
NVLink | 2-Way, 2-Slot, 600 GB/s Bidirectional |
Multi-Instance GPU | Up to 7 MIGs @ 10GB |
Form Factor | PCIe Dual-slot air-cooled or single-slot liquid-cooled |
Graphics bus | PCIe 4.0 x16 |
Thermal solution | Passive |
vGPU Support | NVIDIA Virtual Compute Server (vCS) |
Questions?
We're here to help.
From offering expert advice to solving complex problems, we've got you covered.
Reviews (0)
FAQ
No FAQ Found
Reviews
There are no reviews yet.