ZeroStack, creators of a self-driving cloud that lets users focus on their businesses, has announced that its Self-Driving Cloud Platform can now detect and provide selective end-user access of NVIDIA GPUs to deliver GPU-as-a-Service. This makes the ZeroStack platform the first solution to offer cloud-based, fine-grained access to GPU services.
ZeroStack’s GPU-as-a-Service capability gives customers powerful features to automatically detect NVIDIA GPUs hosted on multiple physical 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, etc., 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.”