Cache Server Fragment Segmentation for Latency Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Content delivery networks (CDNs) face challenges in efficiently managing cache resources to handle high volumes of content requests from client computing devices, leading to strain on content providers and CDN service providers, particularly due to the need for efficient transmission and cost consideration, as well as latency in delivering requested content.

Innovation Solution

The implementation of a CDN-based system that utilizes cache servers with a resource cache component to segment content into initialization, intermediate, and majority fragments for storage and retrieval, where the initialization fragment is transmitted from memory with low latency, and the remaining fragments from higher-latency storage, optimizing data access and transmission.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If content is stored entirely in high-latency storage media, then storage capacity is sufficient, but content delivery latency increases

Engineering Contradiction:
Improvecontent delivery latencyVSAvoidcache resource management complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent segments content into three distinct fragments: initialization fragment, intermediate fragment, and majority fragment. Each fragment is stored in different storage media with different latency characteristics. The initialization fragment is stored in low-latency memory for immediate retrieval, while intermediate and majority fragments are stored in higher-latency storage media. This segmentation allows the system to deliver content with reduced latency by providing the initialization fragment immediately while retrieving the remaining fragments in parallel.

Inventive Principle:
Principle #1Segmentation

2Speed

If all content fragments are retrieved from low-latency memory, then content delivery speed increases, but cache resource consumption increases

Engineering Contradiction:
Improvecontent transmission speedVSAvoidmemory resource consumption
Core Design Contradiction:
SpeedVSQuantity of substance

Solution Approach 1:

The patent applies local quality by assigning different storage locations to different parts of the content based on their retrieval requirements. The initialization fragment, which is always needed first to begin content delivery, is stored in low-latency memory. The intermediate and majority fragments, which can be retrieved in parallel after initialization begins, are stored in higher-latency storage media. This differentiated storage strategy optimizes both delivery speed and memory resource utilization.

Inventive Principle:
Principle #3Local quality

3Productivity

If content is delivered from CDN cache servers, then transmission efficiency improves, but cache resource strain increases with high request volumes

Engineering Contradiction:
Improvecontent delivery efficiencyVSAvoidcache server resource management
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by pre-segmenting content and storing the initialization fragment in low-latency memory before a content request arrives. When a request is received, the system can immediately transmit the initialization fragment without waiting for other fragments to be retrieved from slower storage. This pre-positioning of critical content data enables faster response times and reduces the strain on cache server resources during high request volumes.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11194719B2Cache optimization
Publication Date: 2021.12.07 AMAZON TECH INC
  • US11194719B2 patent drawing
  • US11194719B2 patent drawing
  • US11194719B2 patent drawing

AI summary

A system and method for management and processing of resource requests at cache server computing devices is provided. Cache server computing devices segment content into an initialization fragment for storage in memory and one or more remaining fragments for storage in a media having higher latency than the memory. Upon receipt of a request for the content, a cache server computing device transmits the initialization fragment from the memory, retrieves the one or more remaining fragments, and transmits the one or more remaining fragments without retaining the one or more remaining fragments in the memory for subsequent processing.