Conversation-Based Policy Distribution for SDN Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In software-defined access networks, nodes face challenges with limited storage capacity for policies, wasteful resource usage from downloading unnecessary policies, and outdated policies due to changing cloud computing demands, leading to errors in data routing.
Innovation Solution
Implementing a conversation-based policy distribution method where nodes selectively download and delete policies based on actual data transfer needs, using a map cache request and policy download triggers to manage policy distribution efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If policies are downloaded to the node for every destination device, then the node can enforce policies for all destination devices, but the node's limited storage capacity is exceeded
Solution Approach 1:
The patent extracts only the necessary policies from the complete policy set and downloads them selectively to the node based on actual data transfer needs. This is achieved through monitoring data transfer requests and identifying which policies are required for specific source-destination device pairs, thereby downloading only those specific policies rather than all policies for all destination devices.
Solution Approach 2:
The patent implements partial action by downloading a subset of policies rather than the complete policy set. The node receives and stores only the portion of policies that are currently needed for active data transfers, avoiding the excess of storing all possible policies. This selective approach maintains policy enforcement capability while respecting storage constraints.
2Reliability
If all policies are downloaded to the node, then complete policy coverage is achieved, but network resources are wasted downloading unnecessary policies
Solution Approach 1:
The patent extracts and downloads only the specific policies that are needed for current data transfer operations. By monitoring data transfer requests between source and destination devices, the system identifies and downloads only the relevant policies from the complete policy set, avoiding the waste of network resources on transferring unnecessary policies.
Solution Approach 2:
The node actively monitors its own data transfer needs and triggers policy downloads only when necessary. This self-service approach allows the node to identify when policies are needed based on actual data transfer requests, and automatically initiate downloads only for those specific policies, rather than receiving continuous or blanket policy updates.
3Device complexity
If policies are stored statically at the node, then storage simplicity is maintained, but policies become outdated as the virtualized network structure changes
Solution Approach 1:
The patent implements dynamic policy distribution where policies are not stored statically but are selectively downloaded and updated based on changing network conditions and data transfer needs. As the virtualized network structure changes, the system monitors these changes and dynamically adjusts which policies are downloaded to the node, ensuring policies remain current without requiring complex manual management.
Solution Approach 2:
The system implements feedback mechanisms where data transfer requests and network structure changes are monitored, and this information feeds back into the policy distribution process. When changes are detected in the network structure or data transfer patterns, the system responds by updating the policy set at the node, ensuring policies remain synchronized with the current network state.
4Reliability
If policies are updated frequently to reflect network changes, then policy accuracy is improved, but storage and processing overhead at the node increases
Solution Approach 1:
The patent extracts and downloads only the specific policies that are needed for current data transfer operations, rather than updating the entire policy set frequently. By identifying and downloading only the relevant subset of policies based on active data transfers, the system maintains policy accuracy while minimizing storage overhead at the node.
Solution Approach 2:
The patent implements partial updates by downloading only the necessary portion of policies rather than complete policy set updates. This approach maintains policy accuracy for active transfers while avoiding the storage overhead of maintaining complete policy sets, as only the required subset is stored at the node.
Data Source
AI summary
This disclosure describes techniques for improving policy distribution in a network. The policy distribution may be improved based on a conversation between devices of the network. The techniques include initiating a policy download trigger based on the conversation. The policy download trigger may specify a particular destination device for a data transfer across the network. Therefore, policies specific to the destination device may be downloaded and/or installed at points of enforcement in the network. In this way, the techniques described herein may improve the efficiency of network resource use.