Automatic Load Balancing for Network Path Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Common load balancing techniques in network devices often lead to port overloading and excessive queue delays due to the lack of consideration for state information such as path, device, or queue states, resulting in inefficient path selection and potential packet reordering.
Innovation Solution
The implementation of automatic load balancing (ALB) techniques that select paths based on observed state attributes like path, device, or queue states, maintaining a mapping of optimal paths for active flows and dynamically reassigned paths upon failure or inactivity to minimize reordering and ensure fast failover.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If common load balancing techniques are used, then path selection is simplified, but port overloading and excessive queue delays occur
Solution Approach 1:
The patent implements feedback mechanisms by continuously monitoring queue depths, path states, and device conditions, then using this information to dynamically adjust load balancing decisions. This feedback loop prevents port overloading by redistributing traffic based on real-time system state, resolving the contradiction between simple path selection and reliable load distribution.
Solution Approach 2:
The system transitions from static load balancing to dynamic adaptation by continuously adjusting path selection based on observed state attributes. The load balancer modifies its behavior in real-time according to changing network conditions, queue states, and path availability, thereby maintaining reliable load distribution without sacrificing operational simplicity.
2Power
If state information is not considered, then load balancing computation is reduced, but inefficient path selection and packet reordering occur
Solution Approach 1:
The patent changes the parameters used for load balancing from simple static metrics to dynamic state-aware parameters including queue depths, path states, and device conditions. By incorporating these additional parameters, the system achieves more accurate path selection and reduced packet reordering while maintaining computational efficiency through optimized state monitoring and selection algorithms.
3Device complexity
If static path assignment is used, then implementation is simpler, but failover latency increases
Solution Approach 1:
The system performs preliminary actions by pre-establishing path mappings and maintaining ready-to-use alternative paths based on current system state. When failover is needed, the system can quickly switch to pre-identified alternative paths without extensive recalculation, thereby reducing failover latency while keeping implementation complexity manageable through structured path management.
4Reliability
If dynamic path reassignment is implemented, then load distribution improves, but packet reordering increases
Solution Approach 1:
The patent applies local quality by implementing different path reassignment strategies for different packet types, flows, or priority levels. Critical packets or those sensitive to reordering maintain their original path assignments, while less sensitive traffic can benefit from dynamic reassignment. This localized approach maintains load distribution balance while minimizing packet reordering impacts.
Data Source
AI summary
Automatic load-balancing techniques in a network device are used to select, from a multipath group, a path to assign to a flow based on observed state attributes such as path state(s), device state(s), port state(s), or queue state(s) of the paths. A mapping of the path previously assigned to a flow or group of flows (e.g., on account of having then been optimal in view of the observed state attributes) is maintained, for example, in a table. So long as the flow(s) are active and the path is still valid, the mapped path is selected for subsequent data units belonging to the flow(s), which may, among other effects, avoid or reduce packet re-ordering. However, if the flow(s) go idle, or if the mapped path fails, a new optimal path may be assigned to the flow(s) from the multipath group.