Hot Reservation for Multi-Interface Network Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In-line deployed Network Function (NF) units, particularly Stateful NF units, face challenges in maintaining high availability and redundancy due to limitations in existing protocols like VRRP, which fail to handle multi-interface systems and timely switchover, leading to increased switchover time, packet loss, and network latency.
Innovation Solution
A system with a '1+1' redundancy scheme where a controlling NF unit and a backup NF unit exchange keep-alive messages, with the backup unit automatically triggering a switchover if the controlling unit fails and vice versa, ensuring timely data synchronization and improved operational status, using Virtual IP addresses for bi-directional traffic management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If VRRP protocol is used for redundancy, then failover capability is provided, but switchover time increases and packet loss occurs
Solution Approach 1:
The patent implements hot reservation where the backup NF unit is pre-configured with all necessary routing tables, forwarding rules, and session state information before failure occurs. This preliminary preparation enables immediate switchover without the delays experienced by VRRP, which must perform protocol-specific state synchronization during failover.
Solution Approach 2:
The patent employs bidirectional keep-alive message exchange between controlling and backup NF units to continuously monitor system health. This feedback mechanism detects failures immediately and triggers switchover without waiting for VRRP's longer detection intervals, reducing both detection time and overall switchover duration.
2Reliability
If VRRP protocol is used for redundancy, then failover capability is provided, but packet loss occurs during switchover
Solution Approach 1:
The backup NF unit maintains pre-synchronized session state, routing tables, and forwarding rules identical to the controlling NF unit before failure. This preliminary synchronization ensures that when switchover occurs, the backup unit can immediately handle packets without loss, unlike VRRP which experiences packet loss during its state synchronization process.
Solution Approach 2:
The patent maintains continuous bidirectional keep-alive messaging and state synchronization between NF units, ensuring that the backup unit remains fully operational and ready to immediately assume traffic handling without interruption. This continuous preparation eliminates the packet loss gap that occurs in VRRP during failover transitions.
3Reliability
If traditional redundancy protocols are used, then backup capability is provided, but network latency increases
Solution Approach 1:
The backup NF unit is pre-configured with complete routing tables, forwarding rules, and session state information before failure occurs. This preliminary preparation eliminates the latency introduced by traditional protocols that must perform state synchronization and configuration loading during failover, enabling immediate traffic resumption without delay.
Solution Approach 2:
Bidirectional keep-alive messages provide real-time feedback on system status, enabling immediate detection and response to failures. This feedback mechanism eliminates the detection and response delays inherent in traditional redundancy protocols, reducing overall network latency during failover events.
4Adaptability or versatility
If multi-interface NF units are deployed, then functionality is improved, but existing protocols fail to handle them properly
Solution Approach 1:
The patent designs the hot reservation mechanism to be protocol-agnostic and interface-agnostic, working with any number of network interfaces and any NF type (Stateful or Stateless). The bidirectional keep-alive messaging and state synchronization approach universally applies to multi-interface configurations without requiring protocol-specific modifications, overcoming the limitations of VRRP and other traditional protocols.
Solution Approach 2:
The patent modifies the redundancy approach by implementing interface-specific keep-alive messages and state synchronization for each interface of the multi-interface NF units. This parameter change allows the system to properly handle multiple interfaces independently, whereas traditional protocols like VRRP were designed for single-interface scenarios and fail to properly manage multi-interface configurations.
Data Source
AI summary
Method, device, and system for providing hot reservation for in-line deployed network functions with multiple network interfaces. A system includes a first Network Function (NF) unit, connected to an ingress router and to an egress router; and a second NF unit, connected to the ingress router and to the egress router. The first NF unit is initially configured as a controlling NF. The second NF unit is initially configured as a backup NF. The two NF units periodically exchange keep-alive messages via the two routers. The second NF unit, operating as the backup NF, automatically triggers a switchover if the second NF unit did not receive a keep-alive message from the first NF unit for at least a pre-defined time-period. Additionally or alternatively, the controlling NF initiates a switchover if the maintenance status parameters of the backup NF are better than those of the controlling NF.