Fault-Tolerant Storage Service for Router State Persistence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Communication networks face challenges in maintaining routing information integrity and resilience due to router failures, which can lead to disruptions and inefficiencies in route selection and data processing.
Innovation Solution
The system employs a fault-tolerant storage service (FTSS) that calculates a key value for data elements, selects K computing elements from X nodes based on this key and a mapping schema, and replicates data across these nodes to ensure persistence and efficient recovery from failures, using a shim layer to facilitate communication and state management between protocol points and siblings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If routing information is stored in a single router, then storage simplicity is maintained, but reliability deteriorates due to router failures
Solution Approach 1:
The patent implements replication of routing information across multiple computing elements (K replicas out of X total elements). When a router fails, the routing information can be retrieved from alternative computing elements, ensuring persistence and reliability without requiring complex distributed storage protocols
Solution Approach 2:
The system pre-calculates and stores routing information in multiple computing elements before any failure occurs. The mapping schema is established in advance, so when a router fails, the system can immediately retrieve routing information from pre-designated alternative elements without complex real-time decision-making
2Reliability
If routing information is replicated across multiple nodes, then reliability is improved, but system complexity increases
Solution Approach 1:
The system uses a mapping schema that maps routing information to computing elements based on numerical parameters (indices). This parameter-based approach simplifies the management of replicated data by providing a deterministic, formula-based method to determine which computing elements store which routing information, eliminating the need for complex peer-to-peer coordination protocols
3Reliability
If K replicas are stored across X computing elements, then fault tolerance is enhanced, but storage overhead increases
Solution Approach 1:
The system stores K replicas out of X total computing elements, where K is chosen to provide adequate fault tolerance without replicating to all elements. This partial replication strategy achieves the necessary reliability (ability to tolerate X-K failures) while minimizing storage overhead compared to full replication across all elements
Data Source
AI summary
In certain embodiments, replicating data elements includes calculating a key value for a data element. The key value is calculated from at least a part of content of the first data element. K computing elements are automatically selected from X computing element nodes according to the key value and a mapping schema. K is a greater than 2 and less than X. The computing element nodes each include computer-readable memory embodied within one or more routers. K replications of the data element are automatically written to the computer-readable memory of the K computing element nodes.