Linear Programming Distributed Video Storage Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional video-on-demand streaming technologies face challenges in managing server bandwidth, storage, and network traffic, leading to high deployment costs and inefficiencies in multimedia storage and retrieval systems.

Innovation Solution

A linear programming-based distributed multimedia storage and retrieval system that partitions multimedia data into segments and optimally stores and retrieves them across a network of servers, using local and remote proxy servers, based on storage size and bandwidth conditions, to minimize deployment costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If distributed servers are deployed to provide cost-effective video-on-demand streaming services, then scalability and network load reduction are improved, but deployment costs associated with server bandwidth, storage, and network traffic increase

Engineering Contradiction:
Improvestreaming capacityVSAvoiddeployment cost
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

The patent divides movies into segments and distributes them across multiple servers according to linear programming optimization. This segmentation allows the system to serve more users simultaneously (improving productivity) while optimizing storage allocation to reduce total deployment costs.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses linear programming to optimize parameters such as storage allocation, bandwidth distribution, and server configuration. By changing these parameters based on demand patterns and cost constraints, the system achieves cost-effective deployment while maintaining high streaming capacity.

Inventive Principle:
Principle #35Parameter changes

2Quantity of substance

If more servers are deployed to increase storage capacity, then movie availability is improved, but network traffic among servers increases

Engineering Contradiction:
Improvestorage capacityVSAvoidnetwork traffic
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The patent performs preliminary optimization of server configuration and storage allocation using linear programming before actual streaming occurs. This preliminary action determines the optimal distribution of movie segments and bandwidth allocation, reducing network traffic during operation while maintaining high storage capacity.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system continuously monitors and adjusts server performance based on feedback from streaming demands. This feedback mechanism allows the system to optimize network traffic patterns while maintaining adequate storage capacity, reducing unnecessary data transmission between servers.

Inventive Principle:
Principle #23Feedback

3Ease of operation

If conventional VoD streaming technologies are used, then ease of operation is maintained, but optimization of server bandwidth and storage costs is insufficient

Engineering Contradiction:
Improveservice availabilityVSAvoidbandwidth usage
Core Design Contradiction:
Ease of operationVSQuantity of substance

Solution Approach 1:

The patent implements self-service optimization where the linear programming system automatically adjusts server bandwidth allocation and storage configuration based on real-time demands. This eliminates manual intervention while optimizing resource usage, maintaining ease of operation through automated management.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS9848213B2Linear programming based distributed multimedia storage and retrieval
Publication Date: 2017.12.19 THE HONG KONG UNIV OF SCI & TECH
  • US9848213B2 patent drawing
  • US9848213B2 patent drawing
  • US9848213B2 patent drawing

AI summary

Video on demand is orchestrated between systems by linear program (LP) based multimedia storage and retrieval. An LP based storage and retrieval system can partition multimedia data into an integral number of segments. Further, the LP based storage and retrieval system can store a first amount of the integral number of segments in a storage device of storage devices of a storage network in response to a determination that a storage size of the storage device satisfies a first condition with respect to the first amount of the integral number of segments. Furthermore, such system can store a segment of a second amount of the integral number of segments in a remote storage device of the remote storage devices in response to a determination that the segment satisfies a second condition with respect to a defined number of remote storage devices of the storage devices that have stored the segment.