Cloud Workflow Scheduling via Task Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The current Network Based Media Processing (NBMP) standard lacks functionality for scheduling workflows or parts of workflows on cloud platforms, which can lead to inefficient use of computational resources and peak traffic issues.
Innovation Solution
A method and apparatus for scheduling NBMP workflows on cloud platforms by obtaining an input workflow, generating a modified workflow by dividing the input media stream into tasks, and scheduling these tasks based on a schedule descriptor that includes schedule type information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If workflows are run task by task or in groups without scheduling on cloud platforms, then real-time processing can be achieved, but computational resources are inefficiently utilized and peak traffic issues occur
Solution Approach 1:
The patent applies preliminary action by introducing schedule descriptors that define execution times and scheduling policies before workflows are actually executed. The workflow manager uses these pre-defined schedules to allocate tasks to cloud platforms in advance, avoiding peak traffic periods and optimizing resource utilization before the actual processing occurs.
Solution Approach 2:
The patent implements dynamics by making the workflow execution flexible and adjustable through schedule descriptors. The scheduling system can dynamically adjust when workflows run based on resource availability, traffic conditions, and user-defined preferences, transitioning from rigid real-time execution to adaptive scheduled execution.
2Ease of operation
If the current NBMP standard functionality is used without scheduling, then workflows can be executed immediately, but resource allocation is inefficient and peak traffic cannot be avoided
Solution Approach 1:
The patent applies self-service by enabling workflows to carry their own scheduling instructions through schedule descriptors embedded in the workflow definition. The workflow manager automatically reads and executes these scheduling instructions without requiring external intervention, making the system both simple to operate and efficient in resource allocation.
3Productivity
If workflows are scheduled on cloud platforms with schedule descriptors, then resource utilization is optimized and peak traffic is avoided, but the system complexity increases
Solution Approach 1:
The patent applies universality by designing a multi-functional schedule descriptor that handles multiple scheduling scenarios (immediate execution, delayed execution, periodic execution) through a single unified structure. This universal approach reduces system complexity by providing a consistent interface for diverse scheduling needs rather than requiring separate mechanisms for each scenario.
4Adaptability or versatility
If workflows are divided into tasks for scheduling, then flexible scheduling becomes possible, but the workflow processing complexity increases
Solution Approach 1:
The patent applies segmentation by automatically dividing workflows into discrete tasks that can be independently scheduled. The workflow manager identifies task boundaries and dependencies, then applies scheduling policies to individual tasks or groups of tasks, enabling flexible scheduling while managing complexity through systematic decomposition.
Data Source
AI summary
There is provided a method for scheduling a Network Based Media Processing (NBMP) workflow on a cloud platform. The method includes obtaining an input workflow including input media stream, generating a modified workflow by dividing the input media stream into one or more tasks, scheduling the one or more tasks on the cloud platform based on a schedule descriptor, which includes schedule type information and processing the modified workflow based on the scheduling of the one or more tasks.