CDN DNS Resolution Using Embedded Requester Data for Edge Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The hierarchical structure of the Domain Name System (DNS) limits the availability of accurate information about client computing devices, leading to the selection of suboptimal edge servers in content delivery networks (CDNs) due to the inability of authoritative DNS servers to determine the proximity of edge servers to the actual client devices.
Innovation Solution
Embedding requestor information into the IP address response, allowing edge servers to extract and associate client computing device information, which is used to generate mappings for predicting the characteristics of client populations, enabling authoritative DNS servers to select edge servers based on predicted locations rather than proximate intermediate DNS servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the DNS hierarchical structure is used to resolve hostnames, then the system maintains structural simplicity and ease of operation, but the availability of accurate client information is limited, resulting in suboptimal edge server selection
Solution Approach 1:
The patent embeds requestor information within the IP address response itself, creating a nested structure where additional data is contained within the existing address format. This allows the DNS system to carry client location information through the resolution hierarchy without adding external communication channels or complex protocols.
Solution Approach 2:
The system performs preliminary analysis of client characteristics by embedding requestor information in advance, enabling authoritative DNS servers to make informed edge server selections based on predicted client locations rather than relying on intermediate server locations. This preparatory data embedding resolves the information asymmetry in the DNS hierarchy.
2Productivity
If edge server selection is based on intermediate DNS server proximity, then the resolution process is simple and fast, but the selected edge server may not be optimal for the actual client device
Solution Approach 1:
The system creates a feedback loop by embedding requestor information in IP addresses, allowing edge servers to extract and analyze client characteristics. This feedback enables continuous improvement of edge server selection by matching clients with optimally located edge servers based on historical and real-time data about client locations and behaviors.
Solution Approach 2:
The patent changes the parameters used for edge server selection from intermediate server location to predicted client location. By extracting and utilizing embedded requestor information, the system dynamically adjusts selection criteria to match actual client characteristics, improving content delivery efficiency and reducing network latency.
3Loss of information
If requestor information is embedded in the IP address response, then client characteristics can be accurately determined, but the IP address structure becomes more complex
Solution Approach 1:
The IP address response serves multiple functions: it identifies the edge server through the subnet portion while simultaneously carrying requestor information in the remaining bits. This multi-functionality allows a single data structure to perform both traditional routing and client characterization without requiring separate communication channels or additional protocol overhead.
Data Source
AI summary
A DNS resolution request for a hostname of a CDN is received. An edge server of the CDN may be identified, which may be associated with a subnet. The subnet is used to generate a response IP address, where the remaining bits of the response IP address may be used to store requestor information (e.g., a requestor IP address). When a client computing device uses the response IP address to access the edge server, requestor information is extracted and associated with client computing device information (e.g., an IP address and/or location, etc.) in an association record. Association records may be used to determine predicted characteristics for devices served by a requestor. When the authoritative DNS server resolves a request from the requestor, such predicted characteristics may be used rather than relying solely on information about a requestor. Thus, an edge server proximate to the predicted location may be returned instead.