Content Asset Resource Bidding for Workload-Aware Tasking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inefficient task assignment in managing content asset resources due to lack of knowledge about the current workload of processors and packagers, leading to overloading and suboptimal resource utilization.
Innovation Solution
A bidding system where taskers request bids from resources like video packagers and processors based on their capabilities and availability, allowing for dynamic task allocation and autonomous network management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a tasker assigns tasks to processors and packagers without knowledge of their current workload, then task assignment is simple and fast, but resource utilization becomes suboptimal and overload occurs
Solution Approach 1:
The system implements feedback by having processors and packagers report their current workload status to taskers. Taskers use this feedback information to make informed task assignment decisions, balancing resource utilization across the network while avoiding overload of individual resources.
Solution Approach 2:
Resources (processors and packagers) autonomously manage their own workload by evaluating incoming task requests against their current capacity and making independent decisions about task acceptance and rejection, eliminating the need for complex centralized scheduling.
2Productivity
If a centralized system manages all task assignments, then resource allocation is optimized, but system complexity and single point of failure increase
Solution Approach 1:
The centralized task management function is segmented and distributed to individual taskers across the network. Each tasker independently manages task assignments within its domain, eliminating the single point of failure while maintaining efficient resource allocation through localized decision-making.
Solution Approach 2:
Individual processors and packagers autonomously evaluate and accept or reject task requests based on their own workload capacity, enabling decentralized self-organization of tasks without requiring a centralized controller, thus improving reliability while maintaining efficiency.
3Productivity
If taskers unilaterally decide which packager to task, then task assignment is fast, but adaptability to resource availability changes is reduced
Solution Approach 1:
Taskers continuously receive feedback from processors and packagers about their current workload and availability status. This feedback mechanism enables taskers to dynamically adapt task assignments to changing resource conditions while maintaining efficient task assignment speed through automated decision-making.
Solution Approach 2:
The task assignment system transitions from static unilateral decisions to dynamic multi-party negotiation. Taskers, processors, and packagers continuously exchange information and adjust assignments in real-time based on changing conditions, enabling the system to adapt to resource availability changes while maintaining operational efficiency.
Data Source
AI summary
Provided are methods and systems whereby a request for a task relating to a content asset can be received. Information related to the task can be transmitted. A plurality of bids can be received in response to the transmitted information. Each bid can originate from a corresponding device such as a video packager. Each bid can represent a network related cost associated with the corresponding video packager accepting the task. A winning bid can be determined from the plurality of bids.