Session Resource Manager for On-Demand Video Queuing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
On-demand systems, such as video-on-demand and video game servers, face challenges during peak usage hours due to limited simultaneous session capabilities, leading to poor user experience as users are unable to obtain requested sessions, requiring manual retries and resulting in frustration and inefficiency.
Innovation Solution
Implementing a session resource manager that automates the acquisition of on-demand sessions by queuing requests and allocating resources efficiently, allowing users to access alternate content while waiting and automatically starting sessions when resources become available, thereby reducing the need for manual retries and improving user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the system capacity is increased to meet peak demand, then user experience is improved, but infrastructure cost increases prohibitively
Solution Approach 1:
The system performs preliminary actions by queuing session requests before resources are available. Instead of immediately rejecting users when resources are unavailable, the system accepts requests in advance, maintains them in a queue, and automatically initiates sessions as resources become available, thereby improving user experience without requiring excess infrastructure capacity.
Solution Approach 2:
The patent introduces an intermediary mechanism (session queue and resource manager) between the user request and the resource allocation. This intermediary manages the timing mismatch between request arrival and resource availability, automatically initiating sessions when resources free up, thus resolving the contradiction between meeting demand and maintaining reasonable infrastructure levels.
2Productivity
If manual retry mechanisms are used by users, then session acquisition is possible, but user frustration increases and efficiency decreases
Solution Approach 1:
The system implements self-service by automatically managing session acquisition without requiring user intervention. The resource manager continuously monitors resource availability and automatically initiates sessions for queued requests, eliminating the need for users to manually retry and reducing both frustration and operational complexity.
Solution Approach 2:
The system establishes a feedback loop where the resource manager continuously monitors resource availability and automatically adjusts session initiation based on current system state. This feedback mechanism ensures sessions are started at optimal moments when resources are available, improving acquisition efficiency without user involvement.
3Reliability
If resources are reserved for queued requests, then session acquisition reliability is improved, but resource availability for new requests decreases
Solution Approach 1:
The resource allocation system is made dynamic rather than static. Resources are not permanently reserved but are allocated based on real-time conditions - queued requests are serviced as resources become available, and the system adapts to changing demand patterns. This dynamic allocation maintains reliability for queued users while preserving flexibility for new requests.
Solution Approach 2:
The system changes the parameter of resource allocation from fixed reservation to time-based conditional allocation. Resources are allocated to queued requests based on timing and availability parameters rather than permanent reservation, allowing the system to maintain reliability for queued users while adapting resource distribution to current system state and new requests.
Data Source
AI summary
Methods and systems to provide and manage on-demand sessions are described herein. A first on-demand session for a client is conducted when resources are available for the first on-demand session. When a notification of termination of the first on-demand session is received, the resources are reserved for the client for a time period. When a second request from the client for a second on-demand session is received during the time period, the second on-demand session is started using the reserved resources.