Method and device for allocating computing resources of computing node and medium

• Main Authors: MENG YANG, YU DIANHAI, ZHOU TI, SHA YANLIN
• Format: Patent
• Language: Chinese; English
• Published: 30.08.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention provides a method and device for allocating computing resources of computing nodes, electronic equipment and a computer readable storage medium, and relates to the technical field of computers, in particular to the technical field of deep learning. According to the implementation scheme, in response to a first tensor request memory resource generated in the process of executing a first machine learning task, an available GPU memory in a computing resource is determined; and based on the size of the memory resource requested by the first tensor, determining whether to allocate the memory resource for the first tensor from the available GPU memory. The computing resource includes a total GPU memory of at least one GPU included by the computing node. 本公开提供了一种用于分配计算节点的计算资源的方法、装置、电子设备和计算机可读存储介质，涉及计算机技术领域，尤其涉及深度学习技术领域。实现方案为：响应于在执行第一机器学习任务的过程中产生的第一张量请求内存资源，确定计算资源中的可用GPU内存；以及基于第一张量所请求的内存资源的大小，确定是否从可用GPU内存中为第一张量分配内存资源。计算资源包括计算节点所包括的至少一个GPU的总GPU内存。