Self-Driving Cloud Provider ZeroStack Unveils GPU-as-a-Service via NVIDIA Hardware

zerostackZeroStack, a self-driving cloud platform provider for enterprises and MSPs automating cloud infrastructure, applications, and operations, has announced the launch of cloud-based, fine-grained access to GPU services from its ZeroStack platform. This selective end-user access of NVIDIA GPUs results in ZeroStack starting to bring this GPU-as-a-Service solution to the market.

ZeroStack’s GPU-as-a-Service capability provides customers with new features to automatically detect NVIDIA GPUs hosted on multiple dedicated servers and make them available within the ZeroStack environment. ZeroStack cloud administrators can then configure, scale, and allow fine-grained access control of GPU resources to end-users.

Users can enable GPU acceleration, deploy new machine learning and deep learning workloads with tools such as TensorFlow, Caffe, et cetera, and provide the applications with dedicated access to multiple GPUs “for order-of-magnitude faster inference latency and user responsiveness.”

“Artificial Intelligence and machine learning applications are quickly becoming commonplace, and GPUs are essential for these solutions,” said Kamesh Pemmaraju, vice president of product management at ZeroStack. “By becoming the first to offer GPU-as-a-Service, one that is controlled by administrators, ZeroStack makes it easy to for end users to access GPU resources from within a simple yet powerful cloud environment.”

With ZeroStack, enterprises and MSPs would be able to simplify operations, reduce costs, and accelerate IT while maintaining full IT control. ZeroStack enables multitenant, multi-cloud, and containerized environments with a future-ready architecture. Founded by senior engineers from VMware, ZeroStack is funded by Formation 8 and Foundation Capital, and is based in Mountain View, California