Sharded Network Device Configuration Without Redundant Data Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network device configurations and status retrieval methods consume significant memory and computing resources due to the maintenance of a data tree that duplicates configuration and status information, leading to inefficiencies in storage and processing.
Innovation Solution
A network device that provides access to configuration and status information through a data model schema without storing a data tree, using mapping information to link schema shards to internal data stores, reducing memory consumption and processor load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a data tree is maintained to store configuration and status information, then users can access network device data through standardized protocols, but memory consumption and computing resources increase significantly
Solution Approach 1:
The patent extracts the data tree component from the system, eliminating the need to maintain a separate data tree in memory. Instead, the system directly accesses configuration and status information from internal data stores through mapping mechanisms, removing the redundant data storage layer while preserving access functionality
Solution Approach 2:
The patent makes the internal data stores serve multiple functions: they directly provide both configuration storage and status information retrieval without requiring a separate data tree structure. The mapping information enables these data stores to fulfill the role previously played by the data tree, eliminating redundant functionality
2Ease of operation
If a data tree is maintained to store configuration and status information, then users can access network device data through standardized protocols, but processor load increases due to synchronization requirements
Solution Approach 1:
The patent removes the data tree maintenance and synchronization processes from the system. By eliminating the data tree component, the complex synchronization operations required to keep the data tree updated with internal data store changes are completely removed, reducing processor load and operational complexity
Solution Approach 2:
The patent merges the data tree functionality directly into the internal data stores. Instead of maintaining a separate data tree that requires synchronization with internal data stores, the system combines these functions by using the internal data stores directly with mapping information to provide data tree-like access without the synchronization overhead
Data Source
AI summary
A network device includes an application programming interface (API) to access (read/write) network device configuration and status. For read requests, the API services the request by reading one or more internal data stores to read out the requested data without first reading a data tree that stores network device configuration and state. For write requests, the API services the request by writing data to one or more internal data stores, without first writing to a data tree that stores network device configuration and state.