Signalling Gateway Message Sequencing During SCTP Failover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current SS7 network migration to IP networks faces challenges in ensuring reliability and message sequencing during failures, particularly in distributed Signalling Gateway (SG) environments, where SCTP associations can fail, leading to message loss or duplication.
Innovation Solution
A method is introduced to control local processes within a Signalling Gateway, involving queuing data messages, delaying acknowledgement of failure messages, and initiating traffic diversion to set up an alternate path, ensuring message sequencing and reliability through correlation identification and heartbeat messages.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traffic diversion is initiated immediately upon receiving a failure message, then system reliability is improved, but message sequencing may be compromised causing message loss or duplication
Solution Approach 1:
The patent applies preliminary action by delaying the acknowledgement of failure messages until after traffic diversion is complete and all pending messages have been successfully transmitted through the alternate path. This ensures that the remote process receives the failure acknowledgement only after it has processed all queued messages, thereby maintaining message sequencing integrity while still achieving reliable failover.
Solution Approach 2:
The patent implements beforehand cushioning by queuing pending messages before initiating traffic diversion and holding them during the transition period. This cushioning mechanism prevents message loss or duplication by ensuring that no messages are acknowledged or discarded until the failover is complete and the system is stable in the new configuration.
2Stability of the object's composition
If acknowledgement of failure message is delayed, then message sequencing is maintained, but system response time increases
Solution Approach 1:
The patent maintains continuity of useful action by initiating traffic diversion immediately upon detecting failure, while simultaneously queuing and transmitting pending messages through the alternate path. The acknowledgement is delayed only for the minimal time required to complete this continuous process, rather than waiting for arbitrary delays or manual intervention, thus minimizing time loss while ensuring sequencing integrity.
3Reliability
If all pending messages are transmitted before acknowledgement, then message loss is prevented, but device complexity increases
Solution Approach 1:
The patent implements feedback by establishing a confirmation mechanism where the local process receives acknowledgement from the remote process that all queued messages have been successfully received and processed. This feedback loop allows the system to automatically confirm successful failover and clear message queues, reducing the need for complex manual tracking and verification mechanisms while ensuring no message loss.
Data Source
AI summary
A method of controlling a local process that forms part of a first processing entity, where the first processing entity maintains a plurality of associations with a plurality of remote processes in a second processing entity, comprises receiving a failure message from a remote process indicating a fault affecting an association linking the local process with that remote process; queueing data messages destined for that remote process; controlling the transmission of an acknowledgement of the failure message so that data messages pending on the association are received at that remote process before the acknowledgment of the failure message; and initiating a traffic diversion to set up an alternate path between the first processing entity and the second processing entity for queued data messages.