Autonomous Intranet Load Balancer Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center security measures, such as multiple active data centers, can lead to networking issues like session persistence and route convergence problems, particularly during network topology changes or failures, which can result in service disruptions and client awareness of network problems.
Innovation Solution
An autonomous intranet system that uses interior gateway protocol (IGP) to route traffic between two data centers, each with a load balancer advertising a single-host IP address, allowing for seamless failover and maintaining continuous service by performing health checks and adjusting routing accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple active data centers advertise the same block of IP addresses using BGP, then redundancy and disaster recovery capability are improved, but session persistence deteriorates due to route changes during client sessions
Solution Approach 1:
The patent segments the IP address space by assigning unique IP addresses to each data center instead of advertising the same block from multiple centers. This segmentation allows each data center to have its own distinct routing identity, preventing route flapping that disrupts session persistence while maintaining redundancy through multiple active centers.
Solution Approach 2:
The patent introduces a mediation mechanism where data centers do not directly advertise the same routes to the internet, but rather use a coordinated approach where only the active data center advertises routes at any given time. This intermediary control prevents simultaneous route advertisements that cause routing oscillations and session persistence issues.
2Reliability
If multiple data centers use BGP with route health injection, then failover capability is improved, but route convergence time increases causing client awareness of network problems
Solution Approach 1:
The patent implements preliminary action by pre-configuring health check mechanisms and route advertisement policies before failures occur. The active data center is pre-designated to advertise routes, and health monitoring is continuously performed, allowing for rapid failover without prolonged route convergence delays that would alert clients to network problems.
Solution Approach 2:
The patent employs feedback mechanisms through continuous health checks that monitor data center status. When a failure is detected, the feedback loop triggers immediate route advertisement changes to the standby data center, significantly reducing route convergence time compared to traditional BGP methods that rely on slower route health injection protocols.
3Measurement precision
If DNS records are refreshed to update routing information, then route accuracy is improved, but service continuity deteriorates due to DNS cache expiration requirements
Solution Approach 1:
The patent implements self-service by enabling data centers to autonomously advertise and withdraw routes based on their own operational status without requiring external DNS interventions. This self-managed routing approach maintains continuous service because clients receive updated routing information through persistent BGP sessions rather than waiting for DNS cache expiration and refresh cycles.
Data Source
AI summary
The present disclosure provides a system and method for an autonomous intranet. The autonomous intranet can include two or more data centers, each provided with a load balancer for advertising a common single-host Internet Protocol address upstream for the respective data centers under certain conditions.