Erasure-Coded CDN Streaming for Server Throughput Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed content delivery systems face challenges in efficiently managing server failures and optimizing resource utilization, such as communication bandwidth and storage space, particularly in Content Delivery Networks (CDN) that rely on complex synchronization techniques and expensive infrastructure.
Innovation Solution
The implementation of fractional-storage CDN servers that store erasure-coded fragments and use a fragment pull protocol to dynamically adjust delivery throughput, allowing for immediate compensation by other servers when one server fails, ensuring seamless content delivery without affecting ongoing presentations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If replicas or redundancy encoding are used to deal with server fault, then system reliability is improved, but device complexity and storage requirements increase
Solution Approach 1:
The patent segments content into multiple fragments and distributes them across different servers using erasure coding. Each server stores only a portion of the encoded fragments, reducing individual server storage requirements while maintaining system reliability through distributed redundancy. This segmentation approach resolves the contradiction by dividing the redundancy burden across multiple components rather than requiring full replicas on each server.
Solution Approach 2:
The system dynamically adjusts fragment delivery throughput parameters based on server availability and system load. When servers fail or throughput decreases, the system modifies delivery parameters to compensate, maintaining reliability without requiring static over-provisioning of resources. This dynamic parameter adjustment reduces the need for excessive redundancy while preserving system reliability.
2Stability of the object's composition
If inter-server communication techniques are used to synchronize multiple servers, then content delivery consistency is improved, but device complexity and resource utilization decrease
Solution Approach 1:
Assembling devices perform self-service by autonomously selecting available servers and retrieving fragments without requiring complex inter-server synchronization protocols. The system maintains consistency through the mathematical properties of erasure coding rather than through active coordination between servers, reducing device complexity while preserving content delivery consistency.
Solution Approach 2:
The patent introduces a control server as an intermediary that manages fragment distribution and server coordination. This centralized mediator simplifies the synchronization complexity by providing a single point of control rather than requiring peer-to-peer coordination between all servers, reducing overall device complexity while maintaining consistency.
3Reliability
If dedicated communication lines are used to transport content, then content delivery reliability is improved, but loss of substance (bandwidth cost) increases
Solution Approach 1:
The system uses standard public Internet infrastructure for content delivery rather than requiring dedicated communication lines. Erasure coding enables reliable delivery over unreliable public networks by providing error correction capabilities, making the communication infrastructure multi-functional (handling both standard traffic and error-prone streaming content). This resolves the contradiction by eliminating the need for expensive dedicated lines while maintaining reliability through coding techniques.
4Speed
If large amounts of storage are used to cache content at edges, then content delivery speed is improved, but device complexity and loss of substance (storage cost) increase
Solution Approach 1:
The patent segments content into fragments and distributes them across multiple edge servers with fractional storage capacity. Each server stores only a portion of the total content fragments, reducing individual storage requirements while maintaining fast delivery through distributed access. This segmentation resolves the contradiction by enabling speed through distribution rather than requiring large centralized storage at each edge location.
Solution Approach 2:
The system stores more fragments than strictly necessary for any single content item across the distributed network, creating excessive redundancy at the system level rather than at individual servers. This partial storage approach at each node translates to complete content availability system-wide, enabling fast delivery without requiring each edge server to cache entire content libraries.
Data Source
AI summary
Fault tolerance in a distributed streaming system including fractional-storage CDN servers storing erasure-coded fragments encoded with a redundancy factor greater than one from segments of streaming contents. Each server delivers fragments, at a certain fragment delivery throughput, to multiple assembling devices using a fragment pull protocol, wherein a reduction in the fragment delivery throughput of one of the servers triggers a process in which at least some of the other servers approximately immediately increase their fragment delivery throughput as a reaction to the fragment pull protocol, to compensate for the reduced throughput.