IIoT Data Center Scheduling for Balanced Sub-Platform Workloads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial Internet of Things (IIoT) data sub-platforms face imbalances in computing resource utilization, with some lacking sufficient resources while others have idle resources, leading to inefficiencies in operational demands.
Innovation Solution
A system and method for computation scheduling that involves a management platform to split services across data sub-platforms, determining optimal resource allocation and distributing data packages based on operational and service features, ensuring balanced utilization of computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If computing operations are handled locally on each data sub-platform, then operational autonomy is maintained, but computing resource imbalance occurs between sub-platforms
Solution Approach 1:
The patent segments computing operations into service templates that can be independently scheduled across different data sub-platforms. Each service template is divided into executable units that can be distributed to appropriate sub-platforms based on current resource availability, enabling both operational autonomy and balanced resource utilization.
Solution Approach 2:
The patent introduces a scheduling system as an intermediary between service templates and data sub-platforms. This scheduling system receives service templates, determines optimal execution platforms based on operational features and resource availability, and coordinates resource allocation to balance computing loads across sub-platforms while maintaining operational autonomy.
2Productivity
If computing resources are dynamically allocated based on operational demand, then resource utilization efficiency improves, but system complexity increases
Solution Approach 1:
The patent changes key parameters for resource allocation decisions, including operational features (data volume, processing requirements), service features (resource demands, execution characteristics), and platform states (current resource availability, load levels). By dynamically adjusting these parameters and using predefined scheduling rules, the system achieves efficient resource allocation without excessive complexity.
Solution Approach 2:
The patent performs preliminary actions by pre-defining service templates with their resource requirements and pre-establishing scheduling rules for resource allocation. This preparation allows the system to make quick, efficient decisions during actual operation without complex real-time calculations, reducing overall system complexity while maintaining high resource utilization efficiency.
Data Source
AI summary
Provided are a method, a system, and a storage medium for computation scheduling based on an IIoT data center. The system includes a management platform, the IIoT data center, and a plurality of data sub-platforms in communication with each other. The management platform is configured to. obtain operational features of the plurality of data sub-platforms; determine a first sub-platform and one or more second sub-platforms; determine at least one service to be split based on the operational features and service features corresponding to a plurality of services of the first sub-platform; split the at least one service to be split into a plurality of sub-services based on at least one service feature corresponding to the at least one service to be split; determine processing resource demands corresponding to the plurality of sub-services; and determine a service adjustment parameter based on the processing resource demands and the operational features.