Adaptive Stateless Load Balancing via Flow Table Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Heterogeneous access networks face scalability limitations and challenges in independent scaling and optimization of subscriber access and traffic management, leading to restricted load balancing and resiliency in service provider core networks.
Innovation Solution
Implementing stateless load balancing with an enhanced hash function that redistributes packet flow hashes to active servers upon failure, using a flow table for temporary mapping and restoring mappings upon server recovery, to ensure continuous service and improved resiliency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If stateless load balancing is used to distribute packet flows among servers, then load distribution efficiency is improved, but flow affinity is lost causing session disruption during server failures
Solution Approach 1:
The system dynamically adjusts between stateless and stateful operation modes based on server health status. During normal operation, it uses stateless load balancing for efficiency. Upon detecting server failure, it transitions to stateful flow table lookups to maintain flow affinity, then restores stateless operation when servers recover.
Solution Approach 2:
The system changes the operational parameter from pure stateless hashing to stateful flow table mapping when server failures occur. This parameter change enables the system to maintain flow affinity temporarily while preserving the ability to return to efficient stateless operation.
2Reliability
If flow table is used to maintain flow affinity during server failures, then session continuity is improved, but system complexity increases
Solution Approach 1:
The system segments the flow table into temporary failure-mitigation entries rather than a permanent stateful structure. This segmentation allows flow affinity to be maintained only when needed (during server failures) while keeping the overall system predominantly stateless and simpler.
Solution Approach 2:
The system performs preliminary actions by pre-distributing hash indexes to active servers before failures occur. When failures happen, the pre-configured flow table entries enable immediate flow affinity maintenance without complex real-time decision-making.
3Reliability
If hash indexes are redistributed to active servers upon failure, then load balancing resiliency is improved, but flow affinity is disrupted
Solution Approach 1:
The flow table acts as an intermediary between the redistributed hash indexes and the active servers. It temporarily stores flow-to-server mappings that preserve flow affinity even after hash index redistribution, mediating between resiliency requirements and productivity needs.
4Ease of operation
If access gateway couples subscriber access and traffic management, then service management is simplified, but independent scaling is restricted
Solution Approach 1:
The system segments traffic management functions into dedicated service nodes that are decoupled from access gateways. This segmentation allows independent scaling of service nodes while access gateways maintain simplified subscriber access management.
Solution Approach 2:
Service nodes act as intermediaries between access gateways and core network services. They receive traffic management requests from access gateways and handle load balancing, enabling access gateways to remain simple while providing sophisticated traffic management capabilities.
Data Source
AI summary
In general, techniques are described for load balancing, with a service node, packet flows using stateless load balancing that adapts to server failure to provide flow affinity to initially selected servers for the duration of respective flows. In one example, service node device applies stateless load balancing to packet flows to distribute the flows among a plurality of servers. The service node determines a failure of a failed server and then receives an initial packet of a packet flow from the packet flows and forwards the initial packet to an active server. The service node generates a mapping of the packet flow to the active server, determines a recovery of the failed server, receives a subsequent packet of the packet flow, and forwards the subsequent packet of the packet flow to the active server based at least on the mapping of the packet flow to the active server.