Edge Device Flow Table Service Routing in Single-Layer Data Centers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for efficient methods and apparatus to provide services in data centers with a single-layer architecture, where multiple edge devices are interconnected by multiple links, and existing solutions fail to effectively manage service delivery and data traffic routing.
Innovation Solution
A processor-readable medium stores code that enables edge devices to receive data units with specific characteristics, identify associated service modules using a flow table or hash functions, and send data to these modules for service execution, while also managing service module utilization and load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single-layer architecture is used to collapse the traditional three-layer network structure, then device complexity and network hierarchy are reduced, but efficient service delivery and traffic routing become more difficult to manage
Solution Approach 1:
The patent segments service delivery management by introducing flow tables that separate traffic routing logic from the collapsed network architecture. Edge devices maintain flow tables that segment and organize service rules, allowing complex service management to be broken down into manageable flow entries without requiring multiple network layers.
Solution Approach 2:
The patent introduces flow tables as an intermediary mechanism between the collapsed network layers. These flow tables act as mediators that manage service delivery and traffic routing in the single-layer architecture, providing the necessary control plane functionality without requiring traditional multi-layer hierarchical structures.
2Productivity
If flow tables are used to route data units to service modules, then service delivery efficiency is improved, but memory usage and table management complexity increase
Solution Approach 1:
The patent extracts only the essential routing information needed for service delivery into flow table entries. Instead of storing complete service module details, the flow tables contain extracted key identifiers and routing parameters, reducing memory consumption while maintaining efficient service delivery capability.
Solution Approach 2:
The flow tables store partial information about service modules - only the essential routing keys and identifiers needed for fast lookup. This partial action approach stores just enough data to enable efficient routing without the overhead of complete service module descriptors, optimizing the balance between efficiency and memory usage.
3Productivity
If dynamic service module selection is implemented for load balancing, then service handling capacity is improved, but decision-making time and processing overhead increase
Solution Approach 1:
The patent implements preliminary action by pre-computing and caching service module selection criteria in the flow tables. Load balancing decisions are based on pre-established rules and cached state information, allowing rapid dynamic selection without real-time computation overhead. Service modules are pre-registered and their capabilities are cached for fast lookup.
Solution Approach 2:
The system implements feedback mechanisms where service module performance metrics are continuously monitored and fed back into the flow table decision logic. This feedback enables dynamic load balancing decisions to be made quickly based on current system state, improving service handling capacity while keeping decision-making time minimal through rule-based rather than optimization-based selection.
Data Source
AI summary
In some embodiments, a non-transitory processor-readable medium stores code representing instructions to be executed by a processor. The code causes the processor to receive, at an edge device, a first data unit having a characteristic. The code causes the processor to identify, at a first time, an identifier of a service module associated with the characteristic in response to each entry from a set of entries within a flow table not being associated with the characteristic. The code causes the processor to define an entry in the flow table associated with the characteristic and the identifier of the service module. The code causes the processor to send the first data unit to the service module. The code causes the processor to receive, at the edge device, a second data unit having the characteristic, and send the second data unit to the service module based on the entry.