Google Cloud Expands Collobaration with NVIDIA To Advance AI Computing, Software and Services By CIOTechOutlook Team

Google Cloud Expands Collobaration with NVIDIA To Advance AI Computing, Software and Services

CIOTechOutlook Team | Wednesday, 30 August 2023, 02:42 IST

Google Cloud Next — Google Cloud and NVIDIA unveiled new AI infrastructure and tools that will enable users to quickly and easily construct and deploy large-scale models for generative AI.

Google Cloud CEO Thomas Kurian and NVIDIA founder and CEO Jensen Huang spoke in a fireside chat at Google Cloud Next about how the partnership is bringing end-to-end machine learning services to some of the biggest AI customers in the world, including by making it simple to run AI supercomputers with Google Cloud offerings based on NVIDIA technologies. The Google DeepMind and Google research teams have been using the same NVIDIA technology for the past two years in their new hardware and software integrations.

“We’re at an inflection point where accelerated computing and generative AI have come together to speed innovation at an unprecedented pace,” Huang said. “Our expanded collaboration with Google Cloud will help developers accelerate their work with infrastructure, software and services that supercharge energy efficiency and reduce costs.”

“Google Cloud has a long history of innovating in AI to foster and speed innovation for our customers,” Kurian said. “Many of Google’s products are built and served on NVIDIA GPUs, and many of our customers are seeking out NVIDIA accelerated computing to power efficient development of LLMs to advance generative AI.”

PaxML, a framework developed by Google for creating enormous large language models (LLMs), has been upgraded for NVIDIA accelerated computation.

PaxML now enables developers to employ NVIDIA® H100 and A100 Tensor Core GPUs for sophisticated and fully customizable experimentation and scale. PaxML was originally developed to span multiple Google TPU accelerator slices. The NVIDIA NGCTM software catalog currently offers a PaxML container that has been GPU optimized. Additionally, PaxML is powered by JAX, which has been GPU-optimized using the OpenXLA compiler.