Google becomes the first cloud provider to use Nvidia’s Turing T4 tech

Clare Hopping

14 Nov, 2018

Google has become the first cloud provider to offer customers access to Nvidia’s T4 GPU, helping them test out a cheaper alternative to its high-performance computing (HPC)-focused V100.

Although Google is the first cloud provider to offer the GPU, it’s already seen widespread adoption across physical servers, such as those offered by Dell EMC, Hewlett Packard Enterprise, IBM, Lenovo and Supermicro. In total, it’s available in 57 separate server designs, Nvidia said, making it the most used server GPU it’s ever developed.

“We have never before seen such rapid adoption of a datacenter processor,” said Ian Buck, vice president and general manager of Accelerated Computing at NVIDIA. “Just 60 days after the T4’s launch, it’s now available in the cloud and is supported by a worldwide network of server makers. The T4 gives today’s public and private clouds the performance and efficiency needed for compute-intensive workloads at scale.”

The reason the T4 GPU is so popular is that it’s so powerful. AI, machine learning and even just data processing tasks are carried out at such a scale in the modern datacentre, a powerful GPU is required to handle them. Nvidia’s newest GPU features Turing Tensor Cores and new RT Cores that significantly reduce latency and speed up the processes when used alongside accelerated containerized software stacks.

“Real-time visualization and online inference workloads need low latency for their end users. We are delighted to partner with NVIDIA to offer T4 GPU support for Google Cloud customers,” said Damion Heredia, senior director of Product Management at Google Cloud.

“NVIDIA T4 GPUs for Google Cloud offer a highly scalable, cost-effective, low-latency platform for our ML and visualization customers. Google Cloud’s network capabilities together with the T4 offering enable customers to innovate in new ways, speeding up applications while reducing costs.”