Global Load Balancer Automatic Discovery Across Cloud Data Centers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing network systems face challenges in automatically discovering, scaling, and load balancing virtual service instances across multiple cloud data centers, leading to inefficiencies in managing dynamic service demands and maintaining optimal network performance.

Innovation Solution

Implementing a global load balancing (GLB) device that automatically discovers virtual service instances, receives configuration information from an SDN controller, and applies load balancing algorithms to network traffic, enabling dynamic scaling and load balancing across multiple cloud data centers through integration with DNS and BGP protocols.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual configuration methods are used for load balancing across multiple cloud data centers, then system complexity is reduced, but automation capability and response time to dynamic service demands deteriorate

Engineering Contradiction:
Improveautomatic discovery and load balancingVSAvoidsystem configuration complexity
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The GLB device automatically discovers virtual service instances across multiple cloud data centers by receiving announcements from VSI devices and querying DNS servers without manual configuration. The system self-updates load balancing groups and dynamically adjusts traffic distribution based on real-time service availability and load conditions, eliminating the need for manual intervention while maintaining manageable complexity through automated procedures.

Inventive Principle:
Principle #25Self-service

2Adaptability or versatility

If static load balancing configurations are used, then system stability is improved, but adaptability to dynamic service demands and scaling capability deteriorate

Engineering Contradiction:
Improvedynamic scaling and load balancingVSAvoidconfiguration stability
Core Design Contradiction:
Adaptability or versatilityVSStability of the object's composition

Solution Approach 1:

The system implements continuous feedback mechanisms where GLB devices monitor service availability, traffic load, and performance metrics from virtual service instances across cloud data centers. Based on this feedback, the GLB device dynamically updates load balancing groups and adjusts traffic distribution strategies, enabling automatic adaptation to changing service demands while maintaining operational stability through controlled, procedure-based updates.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The load balancing configuration transitions from static to dynamic through automated discovery and updates. Virtual service instances are automatically added or removed from load balancing groups based on real-time availability and load conditions, allowing the system to adapt its composition dynamically while maintaining stability through standardized procedures and continuous monitoring.

Inventive Principle:
Principle #15Dynamics

3Productivity

If centralized control is implemented across multiple cloud data centers, then coordination and load balancing efficiency are improved, but communication overhead and system response time deteriorate

Engineering Contradiction:
Improveload balancing efficiencyVSAvoidcommunication and coordination time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The centralized control function is segmented and distributed across multiple GLB devices, each responsible for specific cloud data centers or virtual service instances. This segmentation allows parallel processing of load balancing decisions, reducing communication overhead and response time while maintaining coordinated control through standardized procedures and information exchange between GLB devices.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS10567288B1Automatic discovery, scaling, and load balancing of multiple cloud data centers in a software-defined network environment
Publication Date: 2020.02.18 JUNIPER NETWORKS INC
  • US10567288B1 patent drawing
  • US10567288B1 patent drawing
  • US10567288B1 patent drawing

AI summary

In general, techniques are disclosed for automatic discovery and load balancing of virtual service instances of a plurality of cloud data centers within a Software Defined Networking (SDN) or a Network Functions Virtualization (NFV) environment. In one example, a global load balancing device (GLB) of a first cloud data center receives, from an SDN controller, address information for a first set of virtual service instances provided by the first cloud data center and a hostname of a domain for which to perform load balancing across the plurality of cloud data centers. The GLB device requests, from a domain name server (DNS) for the domain, address information for other sets of virtual service instances provided by other cloud data centers. Further, the GLB device applies a load balancing algorithm to direct network traffic to one or more of the virtual service instances provided by the plurality of cloud data centers.