DNS Client Subnet Efficiency via Equivalence Class Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional client-subnet systems face inefficiencies due to the inability to effectively aggregate related IP blocks, leading to increased network, processing, and memory resource usage, especially in real-world server operations, as they cannot aggregate non-adjacent blocks without incidentally including unrelated ones.
Innovation Solution
Implementing equivalence class aggregation in DNS systems, where clients are grouped based on predetermined criteria such as geographical location, network topology, or subscriber IDs, allowing for aggregation of arbitrary CIDR blocks into a single equivalence class, using a representative CIDR block for queries and caching answers for efficient resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional client-subnet systems use granular CIDR blocks for DNS queries, then they can maintain precise geographical and topological separation, but they cannot effectively aggregate non-adjacent blocks leading to increased network, processing, and memory resources
Solution Approach 1:
The patent merges multiple non-adjacent CIDR blocks into a single equivalence class represented by a representative CIDR block. This allows clients from different non-adjacent subnets (e.g., 10.0.0.0/25 and 10.0.1.0/25) to be grouped together and share the same cache entries, reducing the total number of cache entries while maintaining accurate geographical and topological separation through the equivalence class mechanism.
2Use of energy by stationary object
If traditional client-subnet systems use smaller prefix lengths to improve efficiency, then CPU and memory usage decrease, but the inability to aggregate unrelated blocks incidentally included in larger blocks increases network bandwidth consumption
Solution Approach 1:
The patent segments the cache structure by organizing entries into equivalence classes rather than individual CIDR blocks. Each equivalence class is represented by a single representative CIDR block, allowing the system to use larger prefix lengths efficiently. This segmentation enables aggregation of non-adjacent blocks without including unrelated blocks, reducing network bandwidth consumption while maintaining low CPU and memory usage through the representative block mechanism.
3Quantity of substance
If traditional systems aggregate adjacent CIDR blocks into larger subnets, then they can reduce the number of cache entries, but they cannot include non-adjacent blocks without incidentally aggregating unrelated blocks
Solution Approach 1:
The patent introduces an intermediary mechanism - the equivalence class with its representative CIDR block - that mediates between non-adjacent CIDR blocks. This intermediary allows the system to aggregate multiple non-adjacent blocks (e.g., London subnets) without including unrelated blocks (e.g., Bermuda subnet) in the aggregation, maintaining geographical and topological accuracy while reducing the number of cache entries through the representative block approach.
Data Source
AI summary
A method for improving client subnet efficiency by equivalence class aggregation includes receiving a Domain Name System (DNS) query from a client, determining, based on predetermined class criteria, that the client is associated with an equivalency class, searching a cache associated with the equivalence class for an answer corresponding to the DNS query, and upon locating the answer, serving the answer to the client. Membership in equivalence classes can be adjusted, tuned, or otherwise modified based upon one or more criteria or rules. The adjustment of memberships may be a necessary step in response to changing network conditions. A membership in one or more equivalence classes can be adjusted based on determining changes in service (such as changes in bandwidth, quality of service, client location, and so forth), and/or based on detection of a modification to the network topology, and/or based on received input concerning performance.