Workflow Scheduling via Global-Local Orchestrator Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As device ecosystems grow in complexity, determining the optimal deployment of workflows across diverse devices and domains becomes challenging due to increased data and connectivity complexity, making it difficult to efficiently execute workflows and meet service level objectives (SLOs).
Innovation Solution
A global orchestrator decomposes workflows into portions and uses metaheuristic algorithms to determine optimal execution domains, providing these portions to local orchestrators based on domain capabilities and constraints, while local orchestrators further optimize using heuristic techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If device ecosystems grow in complexity with more data and devices, then the ecosystem capabilities and versatility improve, but the difficulty of determining optimal workflow deployment increases
Solution Approach 1:
The patent segments the workflow deployment problem into multiple hierarchical levels: global orchestrator handles high-level domain selection using metaheuristic algorithms, while local orchestrators handle specific device-level assignments using heuristic algorithms. This segmentation reduces the computational complexity of the overall deployment determination by dividing it into manageable sub-problems at different scales.
Solution Approach 2:
The patent introduces orchestrators as intermediary components between the complex device ecosystem and workflow deployment decisions. These orchestrators abstract the complexity by maintaining ecosystem models and using algorithms to translate ecosystem capabilities into optimal deployment configurations, thereby mediating between ecosystem versatility and deployment simplicity.
2Reliability
If more metaheuristic algorithm executions are performed to optimize workflow deployment, then the deployment optimality improves, but the computational time and resources increase
Solution Approach 1:
The patent segments the optimization process into two stages: metaheuristic algorithms at the global orchestrator level for domain-level optimization, and heuristic algorithms at the local orchestrator level for device-level optimization. This segmentation allows the system to achieve high deployment optimality through multiple algorithm executions while managing computational time by distributing the optimization burden across different levels with appropriate algorithmic complexity.
Solution Approach 2:
The patent applies partial optimization by using metaheuristic algorithms for critical domain selection decisions and heuristic algorithms for less critical device-level assignments. This partial action approach ensures sufficient optimization for key decisions while avoiding excessive computational time on less impactful decisions, achieving a balance between deployment optimality and computational efficiency.
Data Source
AI summary
Techniques described herein relate to a method for deploying workflows. The method may include receiving, by a global orchestrator of a device ecosystem, a request to execute a workflow; decomposing, by the global orchestrator, the workflow into a plurality of workflow portions; executing, by the global orchestrator, a metaheuristic algorithm to generate a result comprising a plurality of domains of the device ecosystem in which to execute the plurality of workflow portions; and providing, by the global orchestrator, the plurality of workflow portions to respective local orchestrators of the plurality of domains based on the result of executing the metaheuristic algorithm.