Reporting method of node computing capability and computing nodes

Abstract

The embodiment of the invention provides a node calculating capability reporting method, a calculating node, a cluster management method and a cluster management node, wherein the node calculating capability reporting method concretely comprises the following steps that a calculating node obtains quality information and quantity information of self hardware resources; the calculating node determines the calculating capability information of the hardware resources according to the quality information and the quantity information of the hardware resources; the calculating node reports the calculating capability information of the hardware resources to the management node. The embodiment of the invention can avoid the problem of calculating capability inaccuracy due to the heterogeneous feature of the calculating node.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a method for reporting node computing capabilities, a computing node, a cluster management method and a cluster management node. Background technique [0002] A computer cluster is a computer system that connects multiple computers (also known as computing nodes) to cooperate and complete computing tasks. These computing nodes are located in the same management domain, which has a unique management policy and provides services to users as a whole. For example, Apache Mesos is a cluster architecture of a distributed system, which can realize high-efficiency parallel computing and massive storage. [0003] The process of distributing computing tasks on a computer cluster to multiple computing nodes may be called task scheduling. The scheduling method of existing tasks is usually as follows: each computing node reports its own number of CPU cores to a scheduler in the cluster, a...

AI Technical Summary

Problems solved by technology

[0004] However, computing nodes with unbalanced computing capabilities inevitably exist in a computer cluster, that is, even when the number of CPU cores is the same, these unbalanced computing nodes may still have different computing capabilities

[0005] In this way, in practice, there will be the following situation: a computing task can be subdivided into multiple subtasks, which can be assigned to multiple computing nodes for concurrent execution. The CPU main frequency of the first computing node in the cluster is 1.5GHz, and the CPU main frequency of the second computing node is 3.0GHz. Obviously, when subtasks of the same size are assigned to the two computing nodes for calculation, the two Computing nodes have different completion times for subtasks, which will cause the completion time of the entire task to be determined by the slowest subtask, losing the original parallel computing capability

[0006] There may be another worse situation: if these subdivided subtasks are not parallel, but have some kind of dependency relationship, for example, the calculation output of one subtask may be the input of another subtask, then , the computing power of the whole task is determined by the slowest subtask

[0007] In summary, due to the unbalanced computing power of the computing nodes in the computer cluster, the computing resources of the computer cluster cannot be fully utilized, which undoubtedly affects the computing efficiency.

Images