Content Mapper for Flexible CDN Storage Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Content delivery networks (CDNs) lack flexibility in adapting to customer needs due to inflexible topology and limited interfaces, resulting in missed opportunities for customized content management and delivery.
Innovation Solution
A method and system for flexible content processing in CDNs with geographically distributed points of presence (POPs), using characterization tags to choose storage resources and policies for content objects, and a content mapper to translate file names and paths into unique identifiers for efficient storage and retrieval across multiple POPs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If CDNs use fixed topology with limited interfaces, then system complexity is reduced, but adaptability to customer needs deteriorates
Solution Approach 1:
The patent implements dynamic resource allocation by allowing customers to specify policies that dynamically select storage resources based on content characteristics. The system transitions from fixed topology to flexible resource selection where storage bricks are chosen based on content type, popularity, and customer-defined policies, enabling the system to adapt to varying customer needs without requiring complex reconfiguration of the entire CDN topology.
Solution Approach 2:
The patent changes the parameters of resource selection by introducing characterization tags and policies that describe content properties and storage requirements. Instead of using fixed mapping rules, the system uses flexible parameter-based selection where content is mapped to storage resources based on changing parameters such as content type, access patterns, and customer-specific criteria, thereby improving adaptability while maintaining manageable complexity through standardized parameter definitions.
2Speed
If CDNs have more POPs distributed geographically, then delivery performance is improved, but resource allocation flexibility deteriorates
Solution Approach 1:
The patent segments the CDN infrastructure into independent storage bricks that can be selectively allocated to different POPs based on content requirements. Each storage brick is an independent unit that can be assigned to specific geographic locations according to customer policies, allowing the system to leverage the speed advantage of distributed POPs while maintaining flexible resource allocation through granular segmentation rather than monolithic resource blocks.
Solution Approach 2:
The patent introduces a content mapper as an intermediary component that mediates between content requests and storage resources. The content mapper translates content identifiers into storage resource selections based on policies and characterization tags, enabling flexible resource allocation across multiple POPs without requiring direct manual configuration. This intermediary layer abstracts the complexity of resource distribution while preserving both delivery performance and allocation flexibility.
3Ease of operation
If CDNs use standardized interfaces, then ease of operation is improved, but customization capability deteriorates
Solution Approach 1:
The patent implements universality by defining a standardized policy-based interface that can handle multiple content types and storage requirements through a single unified mechanism. The policy system serves multiple functions: it defines content characteristics, specifies storage requirements, determines resource selection criteria, and controls allocation rules. This multi-functional policy interface maintains ease of operation through standardization while enabling extensive customization by allowing customers to define policies tailored to their specific needs without requiring multiple specialized interfaces.
Data Source
AI summary
A method and system for flexibly processing content objects is disclosed. A content object is stored on one of a plurality of storage devices. A request is received for the content object. Information corresponding to the requested object (e.g., a unique identifier) is transmitted to one or more content directories. A response is received from a first content directory indicating a location of the content object. The content object is requested from the first location, and at least part of the content object is received.