Cloud Network Balancing via Shared IP and Load-Aware Routing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current cloud computing network resource load balancing methods require external interactions and additional protocols, reducing efficiency by necessitating communication between the cloud computing platform and clients, and involving complex configurations for virtual machines acting as gateways or management entities.

Innovation Solution

A network resource balancing method and apparatus that utilize a balancing module within the cloud computing platform to receive identifier and load information of virtual machines, determine optimal virtual machine selection based on load information, and send access requests directly, eliminating the need for external client interactions and new protocols by using shared IP addresses and distinct MAC addresses for virtual machines and domains.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If network resource load balancing is implemented using external client interactions and new protocols, then virtual machine selection can be achieved, but network resource balancing efficiency is reduced

Engineering Contradiction:
Improvenetwork resource balancing efficiencyVSAvoidprotocol complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces a balancing module as an intermediary component within the cloud computing platform that receives access request packets, determines target virtual machines based on load information, and forwards packets accordingly. This intermediary eliminates the need for complex external protocols by handling load balancing internally through a dedicated module that maintains load information of virtual machines and makes routing decisions based on current system state.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The cloud computing platform performs load balancing autonomously through its internal balancing module without requiring external clients to implement complex selection protocols. The platform self-manages the distribution of access requests by monitoring virtual machine load information and automatically routing packets to appropriate virtual machines, thereby improving efficiency by eliminating external coordination overhead.

Inventive Principle:
Principle #25Self-service

2Productivity

If independent IP addresses are assigned to each virtual machine, then virtual machine identification is enabled, but network resource utilization is reduced

Engineering Contradiction:
Improvenetwork resource utilizationVSAvoidIP address quantity
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent implements a shared IP address mechanism where multiple virtual machines within the same domain share a common IP address. The balancing module receives access request packets containing this shared IP address, determines the target virtual machine based on load information, and forwards the packet accordingly. This universal IP addressing approach allows the same IP address to serve multiple virtual machines dynamically, significantly improving network resource utilization while maintaining the ability to identify and route to specific virtual machines.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS10931745B2Network resource balancing method and apparatus
Publication Date: 2021.02.23 HUAWEI TECH CO LTD
  • US10931745B2 patent drawing
  • US10931745B2 patent drawing
  • US10931745B2 patent drawing

AI summary

A network balancing method and apparatus, where the method includes obtaining, by a processor in a cloud computing platform, identifier information of a virtual machine and identifier information of a domain including virtual machines that are sent by a controller in the cloud computing platform, where a correspondence exists between the identifier information of the virtual machine and the identifier information of the domain including the virtual machines, and after receiving an access request packet, determining, by the processor, the virtual machines in the domain according to the identifier information of the domain in the access request packet, further selecting a virtual machine according to load information of the virtual machines in the domain, and sending the access request packet to the selected virtual machine.