首页|Method And System For Multiple Services To Share Same Gpu, And Device And Mediu m' in Patent Application Approval Process (USPTO 20240095082)

Method And System For Multiple Services To Share Same Gpu, And Device And Mediu m' in Patent Application Approval Process (USPTO 20240095082)

扫码查看
The following quote was obtained by the news editors from the background informa tion supplied bythe inventors: “It has been very popular to provide computing p ower by using Graphics Processing Units(GPUs) in machine learning and deep lear ning. A typical scene is, in a data center, based on Kubernetes(an open-source container arranging engine, used for automated deployment, scaling and managemen tof containerized applications) as the container arranging environment, constru cting a cloud-environmentcluster to deploy the services of the machine learning and the deep learning. The nodes (servers) in thecluster are divided into diff erent types, where the nodes equipped with a GPU are referred to as GPUnodes, a nd the other nodes are CPU nodes. The GPU nodes serve for the particular tasks o f the machinelearning and the deep learning. The CPU nodes serve for cluster ma nagement, service dispatching and soon. However, because the resources provided by a single GPU such as the graphic memories, the registersand the threads are very abundant, usually one Kubernetes Pod (the smallest unit of Kubernetes) cannot completely utilize the resources of the single GPU such as the graphic memor ies, the registers andthe threads. Therefore, a technique is required to realiz e dispatching a plurality of Pods of a plurality ofservices to the same GPU, th ereby realizing a high GPU utilization ratio.”

CyborgsEmerging TechnologiesMachine LearningPatent Application

2024

Robotics & Machine Learning Daily News

Robotics & Machine Learning Daily News

ISSN:
年,卷(期):2024.(Apr.4)