Global Load Balancing Service Coordination for Data Center Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center configurations face challenges with service access interruptions due to network or hardware failures, and single data centers are inadequate for cross-regional and cross-operator access, leading to unsatisfactory performance.
Innovation Solution
A service coordination method and apparatus for global load balancing (GLB) devices that monitor node-pair states, synchronize DNS parsing lists, and determine the best server IP addresses for optimal user access experiences by implementing a state management unit, DNS processing unit, and synchronizing unit within GLB devices across multiple data centers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single data center is used, then device complexity is reduced, but service access speed and cross-regional access performance deteriorate
Solution Approach 1:
The system segments the data center service into multiple regional data centers, each serving specific geographic regions. The GLB device divides DNS parsing lists by data center and distributes them to appropriate regional GLB nodes, enabling localized access that improves speed while maintaining manageable complexity through structured segmentation.
Solution Approach 2:
The patent introduces a new dimension of geographic distribution by deploying data centers across multiple regions rather than concentrating services in a single location. This spatial dimensionality change enables cross-regional access optimization without significantly increasing operational complexity.
2Device complexity
If a single data center is used, then device complexity is reduced, but service access stability deteriorates due to network or hardware failures
Solution Approach 1:
The system segments services across multiple data centers so that failures in one region do not affect others. Each data center operates independently with its own GLB node, providing fault isolation that enhances reliability while keeping individual data center configurations simple and manageable.
Solution Approach 2:
The patent implements prior cushioning by pre-distributing DNS parsing lists to multiple regional GLB nodes before failures occur. This ensures that if one data center fails, other nodes already have the necessary routing information to maintain service availability, cushioning against the impact of failures.
3Reliability
If multiple data centers are used, then service access stability and speed are improved, but device complexity increases
Solution Approach 1:
The patent merges the coordination function into a centralized GLB device that manages DNS parsing list distribution across multiple data centers. This consolidation simplifies the overall system by providing unified management of multiple centers, reducing the complexity that would otherwise arise from distributed coordination among numerous independent nodes.
Solution Approach 2:
The GLB device performs multiple functions including DNS parsing, service status monitoring, and coordinated distribution of routing information across all data centers. This multi-functionality reduces the need for separate specialized systems at each data center, thereby reducing overall device complexity while maintaining the benefits of multiple centers.
4Speed
If multiple data centers are used, then service access speed is improved, but device complexity increases
Solution Approach 1:
The patent applies local quality by distributing DNS parsing lists to regional GLB nodes based on their specific geographic locations and service requirements. Each node receives only the parsing lists relevant to its region, optimizing access speed for local users while keeping each node's configuration simple and focused on its specific function.
Data Source
AI summary
According to an example, a service coordination apparatus may include a processor that is to determine whether each GLB node-pair in a node-pair state table is at a normal state after receiving a DNS request. If a GLB node-pair in the node-pair state table is not at the normal state, a second DNS response may be sent to a node sending the DNS request according to a DNS parsing list. If each GLB node-pair in the node-pair state table is at the normal state, a proximity list may be inquired. If a record corresponding to the DNS request is searched out from the proximity list, a first DNS response may be sent to the node sending the DNS request according to the searched out record. If not, a second DNS response may be sent to the node sending the DNS request according to the DNS parsing list.