NVIDIA’s AI-focused Ampere A100 GPUs Energizes Google Cloud to the Next Level

Jul 10

Nvidia today announced that its new Ampere-based data center GPUs, the A100 Tensor Core GPUs, are now available in alpha on Google Cloud. The speedy time to availability is about only a month from its launch is itself a history in making. At this time, the pricing information has not yet been released.

Nvidia A100 GPU aims to help businesses to do the heavy lifting of AI and scientific computing. There is a great demand out there. The arrival of the Nvidia A100 GPU comes just in time because many cloud users are looking for AI-based solutions to help their computations on areas like genomics, video analytics, and 5 G services.

Why NVIDIA’s A100 is a breakthrough?

First, it has new capabilities such as Tensor Cores, mixed-precision modes, multi-instance GPU, advanced memory management. Second, the A 100 can work in both smaller and large workloads because it can adjust itself to match the specific applications’ needs.

How this NVIDIA GPU in the Google Cloud can help computing in every workload and scale?

As you scale up your workloads, you have a variety of NVIDIA GPU’s available. On the Google Compute Engine, GCE, ranging from P4, T4, K80, P100, V100, and now A100 is the newest GPU generation. In Google cloud, you can you have up to instances of 16 of these A100 GPUs. This means that you have 640 gigabytes and 1.3 terabytes of system memory.

Th new Ampere-based data center GPUs, the A100 Tensor Core GPUs were designed for AI workloads, as well as data analytics and high-performance computing solutions. Nvidia says the A100 can boost training and inference performance by over 20x compared to its predecessors (though you’ll mostly see 6x or 7x improvements in most benchmarks) and tops out at about 19.5 TFLOPs in single-precision performance and 156 TFLOPs for Tensor Float 32 workloads.

“With our new A2 VM family, we are proud to be the first major cloud provider to market Nvidia A100 GPUs, just as we were with Nvidia’s T4 GPUs. We are excited to see what our customers will do with these new capabilities,” said Director of Product Management at Google Cloud. Manish Sainani.

See the V100 vs A100 Performance Comparison