Network Media Workflow Slicing for Peak-Demand Task Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network-based media processing systems lack flexibility in workflow execution modes and task scheduling, leading to inefficient resource utilization and potential resource shortages during peak demand.
Innovation Solution
Implementing mechanisms for 'breakable' workflows and slicing windows, allowing tasks to be separated into allocated and scheduled states, with the ability to bind tasks later, and enabling flexible execution modes such as streaming and stepping, which optimize resource allocation and utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If tasks are pre-allocated and executed in a fixed workflow manner, then workflow execution is straightforward and simple to manage, but resource utilization becomes inefficient and resource shortages occur during peak demand
Solution Approach 1:
The patent implements dynamic workflow execution by introducing slicing windows that can adapt their boundaries and content based on real-time resource availability and demand conditions. Tasks are not rigidly fixed in predetermined windows but can be dynamically allocated, suspended, and resumed across different slicing windows, allowing the system to optimize resource utilization while maintaining operational simplicity through automated management.
2Productivity
If all tasks are allocated and executed simultaneously, then workflow completion speed is maximized, but resource availability becomes insufficient during peak demand periods
Solution Approach 1:
The patent segments the workflow execution into multiple slicing windows, dividing tasks into smaller batches that can be executed sequentially or in parallel across different time periods. This segmentation allows the system to control resource consumption rates, preventing resource exhaustion while maintaining overall workflow progress. Tasks can be distributed across multiple slicing windows based on resource availability, ensuring reliable execution without overwhelming system capacity.
3Adaptability or versatility
If tasks are separated into different slicing windows, then resource allocation becomes more flexible and utilization optimizes, but workflow management complexity increases
Solution Approach 1:
The patent creates a universal workflow management framework that handles multiple execution modes (synchronous, asynchronous, iterative) and task states (allocated, suspended, resumed, completed) through a single set of管理机制. The slicing window structure serves multiple functions: resource allocation, task scheduling, progress tracking, and state management. This multi-functional approach increases adaptability while controlling complexity through standardized management procedures.
4Productivity
If tasks are suspended and resumed later, then resource utilization optimizes during peak demand, but workflow execution time increases
Solution Approach 1:
The patent implements periodic workflow execution through slicing windows that operate in cycles, with tasks suspended and resumed at predetermined intervals or based on resource availability conditions. This periodic action allows the system to batch process tasks during high-resource periods while maintaining progress, and pause during low-resource periods, optimizing overall resource utilization without significantly extending total workflow completion time through efficient resumption mechanisms.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A network apparatus distributes a first plurality of tasks for processing first media content among at least a first slicing window and a second slicing window based on a connection map included in a workflow description from a network based media processing (NBMP) source. The first slicing window includes at least one first task from among the first plurality of tasks and the second slicing window includes at least one second task from among the first plurality of tasks. The network apparatus provisions the first media content to at least a first of the one or more media sinks by deploying the at least one first task to one or more first media processing entities, and deploying the at least one second task to one or more of the first media processing entities in response to receiving an indication that the at least one first task has been deployed successfully.