Virtual BRAS Cluster Redundancy via Common Control Plane
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current single-BRAS solutions lack effective redundancy, leading to service disruptions and visible outages when one BRAS fails, as they rely on limited in-box redundancy and dual-attachment to neighboring LER systems, which cannot compensate or hide complete BRAS outages from end customers.
Innovation Solution
A Virtual Broadband Remote Access Server (BRAS) Cluster with stateful interchassis redundancy (SICR) is implemented, featuring a common control plane and management plane across multiple BRAS units, ensuring seamless switchover of customer sessions and stateful information between BRAS chassis without service disruption, using protocols like LACP and L2TP for synchronization and traffic management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single-BRAS solution is used with limited in-box redundancy, then device complexity is reduced, but reliability deteriorates as complete BRAS outage cannot be compensated
Solution Approach 1:
The invention segments the BRAS functionality into separate virtual machines (control plane, data plane, management plane) that can be distributed across multiple physical BRAS chassis. This segmentation enables redundancy without requiring each individual BRAS to contain all functional components, thus improving reliability while managing complexity through functional separation.
Solution Approach 2:
The invention introduces a virtualization layer as an intermediary between the physical BRAS hardware and the network services. This virtualization platform manages the distribution and failover of virtual machine instances across multiple chassis, providing redundancy while abstracting the complexity from the network operations team.
2Reliability
If dual-attachment towards Backbone is implemented, then link failure protection is improved, but device complexity increases with additional LER connections
Solution Approach 1:
The invention merges multiple BRAS instances into a single virtualized BRAS cluster that presents a unified interface to the backbone network. This consolidation provides redundancy and link failure protection through the virtualized architecture without requiring each physical BRAS to maintain separate dual-attachments, thus reducing overall system complexity while maintaining reliability.
3Reliability
If stateful interchassis redundancy is implemented, then service continuity is improved during failover, but synchronization overhead increases
Solution Approach 1:
The invention implements preliminary action by pre-synchronizing critical session state information between virtual machine instances before failover is needed. The virtualization platform maintains ready-to-use session contexts and configuration data across chassis boundaries, enabling immediate failover without extensive real-time synchronization during actual failure events, thus reducing overhead while ensuring service continuity.
4Reliability
If multiple BRAS chassis are virtualized into a single logical BRAS, then redundancy is improved, but management complexity increases
Solution Approach 1:
The virtualization platform implements self-service capabilities that automatically manage the distribution, monitoring, and failover of virtual machine instances across multiple BRAS chassis. The system autonomously handles load balancing, health checking, and session migration without requiring manual intervention, thus providing enhanced redundancy while simplifying operations through automation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Virtual Broadband Remote Access Server (BRAS) Cluster comprising - at least two BRAS, each of the BRAS comprises a Data Plan, Control Plane, Management Plane; - wherein a common control plane and a common management plane spans over the at least two BRAS, implementing a redundancy on the Virtual BRAS cluster by providing the functionality of the control and management plane, by controlling and managing the at least two BRAS; the common control plane and a common management plane being the external access point to the virtual redundant BRAS, wherein in case of a failure of one of the BRAS the common control and common management plane, switchover the functionality to the working BRAS.