Session Migration Scheduling for CDN Content Delivery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current content delivery network (CDN) service scenarios face inefficiencies in content scheduling, leading to high interaction delays and resource consumption between terminals and servers, particularly in redirection and proxy modes, which fail to effectively manage missed content requests and degrade overall service performance.
Innovation Solution
A session migration-based scheduling method that reduces information exchange between servers and terminals by determining the target IP address for content location through database queries, allowing the first media server to redirect requests to a second media server if it lacks the content, thereby minimizing interactions and resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redirection mode is used for content delivery, then content can be delivered when the first media server misses it, but the interaction delay between terminal and server increases significantly
Solution Approach 1:
The system performs preliminary actions by having the first media server query the database for the target IP address before the terminal needs to be redirected. The database is pre-configured with content-location mappings, allowing the server to proactively determine where content is located and establish connections in advance, rather than waiting for redirection to occur after a miss.
Solution Approach 2:
The patent introduces a database as an intermediary component that stores the mapping relationship between content and server IP addresses. This intermediary enables the first media server to efficiently query content location without direct terminal-server negotiation, reducing interaction delays while maintaining reliable content delivery through informed redirection.
2Reliability
If proxy mode is used for content delivery, then content can be obtained from the second media server, but server resources are consumed and overall service performance degrades
Solution Approach 1:
The patent extracts the content delivery function from the first media server by directly routing terminal requests to the second media server when the first server misses content. Instead of the first server proxying all content requests, the system extracts and directs traffic to the appropriate server based on database queries, reducing resource consumption on the first server while maintaining reliable content delivery.
Solution Approach 2:
The terminal effectively performs self-service by being directly connected to the second media server for content retrieval when the first server misses content. The first media server only needs to provide the target IP address through database querying, and the terminal independently completes the content acquisition from the second server, reducing the service burden on the first server and improving overall service performance.
3Ease of operation
If the first media server directly serves the terminal without session migration, then service simplicity is maintained, but information exchange between server and terminal increases and delays occur
Solution Approach 1:
The database serves as an intermediary that provides the first media server with the target IP address information needed for efficient session migration. This intermediary enables the server to make informed routing decisions without increasing operational complexity, as the database automatically resolves content-location mappings through simple queries.
Solution Approach 2:
The system dynamically migrates sessions between servers based on content availability. When the first media server determines content is not available locally, it dynamically redirects the terminal session to the second media server using the target IP address from the database. This dynamic approach maintains service simplicity while reducing information exchange delays by routing requests to the most appropriate server.
4Measurement precision
If multiple interactions occur between terminal and server for content retrieval, then content location can be determined, but server resource consumption increases
Solution Approach 1:
The database is pre-configured with accurate content-location mappings before any terminal requests occur. This preliminary preparation allows the first media server to query the target IP address in a single operation, determining content location accurately without requiring multiple interaction cycles between the terminal and server, thereby reducing server resource consumption while maintaining precise content location determination.
Data Source
AI summary
A session migration-based scheduling method, where the method includes: receiving a service request from a terminal, where the service request is used to obtain target content required by the terminal; querying a target Internet Protocol (IP) address in a database based on the service request, where the target IP address is an IP address of a server in which the target content is located; determining the target IP address based on a candidate IP address fed back by the database; if the target IP address is different from an IP address of the first media server, determining, by the first media server, that the first media server is missing the target content; and sending the service request to a second media server, where an IP address of the second media server is the target IP address.