Resource processing method of cloud platform, related device, and storage medium

Patent metadata

Field	Value
Publication number	US-11966792-B2
Application number	US-202117159083-A
Country	US
Kind code	B2
Filing date	Jan 26, 2021
Priority date	Jul 22, 2019
Publication date	Apr 23, 2024
Grant date	Apr 23, 2024

Abstract

Embodiments of the present disclosure provide a resource processing method of a cloud platform and a related computing device. The cloud platform includes a plurality of nodes, and each node includes a first resource and a second resource. The method is performed by the computing device and/or a host manager in the cloud platform, and includes: obtaining a current first resource allocation ratio and a current second resource allocation ratio of each of the plurality of nodes; obtaining a first resource limit and a second resource limit that are required by a newly added host; determining, among the plurality of modes, a target node according to the current first resource allocation ratio and the current second resource allocation ratio of each node, the first resource limit, and the second resource limit; and deploying the newly added host in the target node.

First claim

What is claimed is:

1. A resource processing method of a cloud platform, the cloud platform comprising a plurality of nodes, each node comprising a first resource and a second resource different from the first resource, the method being performed by a computing device in the cloud platform, and the method comprising:
obtaining a current first resource allocation ratio and a current second resource allocation ratio of each of the plurality of nodes;
obtaining a first resource limit and a second resource limit that are required by a newly added host;
determining, among the plurality of nodes, a target node according to the current first resource allocation ratio and the current second resource allocation ratio of each node, the first resource limit, and the second resource limit;
deploying the newly added host in the target node;
determining a current state of each node according to the current first resource allocation ratio and the current second resource allocation ratio of the node, the current state indicating whether there is an inter-resource balance between usage of the first resource and the second resource at different nodes and the current state being one of three states: a resource balanced state, a first resource unbalanced state, or a second resource unbalanced state that is opposite to the first resource unbalanced state, wherein the first resource and the second resource are different types of resources; and
migrating a stock host from a node in the first resource unbalanced state that causes first resource allocation imbalance to a node in the second resource unbalanced state or the resource balanced state.

2. The method according to claim 1 , wherein after the obtaining a first resource limit and a second resource limit that are required by a newly added host, the method further comprises:
obtaining a current first resource margin and a current second resource margin of each node, the current first resource margin being the first resource of the entire node minus a sum of partial first resources allocated by the node to stock hosts deployed in the node, and the current second resource margin being the second resource of the entire node minus a sum of partial second resources allocated by the node to the stock hosts deployed in the node;
determining candidate nodes in the plurality of nodes according to the current first resource margin and the current second resource margin of each node, and the first resource limit and the second resource limit that are required by the newly added host; and
selecting target candidate nodes from the determined candidate nodes.

3. The method according to claim 2 , wherein the determining candidate nodes in the nodes according to the current first resource margin and the current second resource margin of each node, and the first resource limit and the second resource limit that are required by the newly added host comprises:
determining a node as a candidate node in the plurality of nodes in accordance with a determination that the current first resource margin of the node is greater than or equal to the first resource limit required by the newly added host, and the current second resource margin of the node is greater than or equal to the second resource limit required by the newly added host.

4. The method according to claim 2 , wherein the selecting target candidate nodes from the determined candidate nodes further comprises:
obtaining a current first resource utilization rate of each candidate node;
obtaining a current first resource utilization rate of the newly added host;
calculating an updated first resource utilization rate of each candidate node according to the current first resource utilization rate of the candidate node and the current first resource utilization rate of the newly added host;
determining the target candidate nodes among the candidate nodes according to the updated first resource utilization rate of each candidate node; and
identifying one target node in the determined target candidate nodes as the target node.

5. The method according to claim 4 , wherein the determining the target candidate nodes among the candidate nodes according to the updated first resource utilization rate of each candidate node comprises:
determining, in accordance with a determination that the updated first resource utilization rate of a candidate node in the candidate nodes is less than or equal to a specified high load threshold, the candidate node as a target candidate node in the candidate nodes.

6. The method according to claim 4 , wherein the determining, among the plurality of nodes, a target node according to the current first resource allocation ratio and the current second resource allocation ratio of each node, the first resource limit, and the second resource limit comprises:
obtaining a current resource imbalance degree of each target candidate node according to the current first resource allocation ratio and the current second resource allocation ratio of the target candidate node, the current resource imbalance degree being used for measuring a degree of difference between the first resource allocation ratio and the second resource allocation ratio of the target candidate node;
calculating an updated first resource allocation ratio and an updated second resource allocation ratio of each target candidate node according to the first resource limit and the second resource limit;
calculating an updated resource imbalance degree of each target candidate node according to the updated first resource allocation ratio and the updated second resource allocation ratio of the target candidate node, the updated resource imbalance degree being used for measuring a degree of difference between the updated first resource allocation ratio and the updated second resource allocation ratio of the target candidate node; and
determining the target node according to the current resource imbalance degree and the updated resource imbalance degree of each target candidate node.

7. The method according to claim 6 , wherein the determining the target node according to the current resource imbalance degree and the updated resource imbalance degree of each target candidate node comprises:
obtaining a resource imbalance alleviation degree of each target candidate node according to the current resource imbalance degree and the updated resource imbalance degree of the target candidate node; and
using a target candidate node having the highest resource imbalance alleviation degree as the target node of the newly added host.

8. The method according to claim 1 , wherein the migrating a stock host from a node in the first resource unbalanced state that causes first resource allocation imbalance to a node in the second resource unbalanced state or the resource balanced state comprises:
calculating a first resource imbalance degree of each node in the first resource unbalanced state;
determining a first resource node having the highest first resource imbalance degree; and
migrating a stock host from the first resource node that causes first resource allocation imbalance to a node in the second resource unbalanced state.

9. The method according to claim 8 , wherein the migrating a stock host from the first resource node that causes first resource allocation imbalance to a node in the second resource unbalanced state comprises:
determining, in accordance with a determination that a ratio of a first resource quota to a second resource quota of a stock host in the first resource node is greater than a ratio of the current first resource allocation ratio to the current second resource allocation ratio of the first resource node, the stock host as a stock host that causes first resource alloc

Assignees

• Tencent Tech Shenzhen Co Ltd

Inventors

• Chen Xinquan

Classifications

• G06F9/5077 Primary

  Logical partitioning of resources; Management or configuration of virtualized resources (specific details on emulation or internal functioning of virtual machines G06F9/455) · CPC title

• G06F9/505

  considering the load · CPC title

• G06F9/5072

  Grid computing · CPC title

• G06F9/5083 Primary

  Techniques for rebalancing the load in a distributed system · CPC title

• G06F2209/5022

  Workload threshold · CPC title