Intelligent Scheduling in Hybrid Cloud-Edge Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud computing systems face service delays and security issues due to inefficient data processing and transmission, particularly in hybrid cloud-edge environments, where resource optimization and collaboration are not adequately addressed.
Innovation Solution
An intelligent scheduling apparatus and method that configures schedulers in a hybrid cloud environment to optimize vertical and horizontal collaboration between cloud, edge, and near-edge systems, using series and parallel connections to process tasks efficiently, manage resource allocation, and train scheduler policies based on historical data for improved performance and latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If data is processed by a centralized data center in cloud computing, then data processing capacity is improved, but service delay increases due to transmission distance
Solution Approach 1:
The patent segments the centralized cloud data center into distributed edge computing nodes deployed closer to users. This segmentation allows data processing to occur locally at edge nodes rather than requiring long-distance transmission to centralized data centers, thereby reducing service delay while maintaining processing capacity through distributed computation.
Solution Approach 2:
The patent introduces a spatial dimension to data processing by deploying edge computing nodes across multiple geographic locations rather than concentrating all processing in a single centralized data center. This dimensional distribution enables users to access nearby edge nodes, reducing transmission distance and service delay while preserving overall processing capacity.
2Ease of manufacture
If resources are separately deployed for respective execution types in conventional edge systems, then service provision is simplified, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent implements universal resource pools at edge nodes that can dynamically serve multiple execution types and service scenarios. Instead of dedicating separate resources for each execution type, the system creates multi-functional resource pools that can be flexibly allocated based on real-time demands, thereby improving resource utilization efficiency while maintaining service provision simplicity through unified management interfaces.
Solution Approach 2:
The patent introduces dynamic resource allocation mechanisms that allow edge computing resources to adaptively adjust their allocation based on changing service demands and execution types. This dynamic approach enables the same physical resources to serve different functions at different times, improving overall resource utilization efficiency while preserving the simplicity of service provision through automated orchestration.
3Device complexity
If only partial vertical collaboration is provided in conventional edge systems, then system complexity is reduced, but collaboration effectiveness deteriorates
Solution Approach 1:
The patent implements comprehensive vertical collaboration mechanisms with feedback loops that enable effective coordination between edge nodes, core network, and cloud platforms. The system collects performance data from edge services and uses this feedback to optimize resource allocation, task scheduling, and service delivery across the entire vertical stack, thereby improving collaboration effectiveness without excessively increasing system complexity through automated control algorithms.
Solution Approach 2:
The patent merges previously separate management functions into an integrated vertical collaboration framework that coordinates edge computing, network transmission, and cloud processing as a unified system. This consolidation improves collaboration effectiveness by eliminating silos and enabling seamless end-to-end optimization, while managing system complexity through standardized interfaces and protocols.
4Ease of operation
If cloud servers schedule tasks based on pre-set time periods, then scheduling simplicity is improved, but adaptability to real-time conditions deteriorates
Solution Approach 1:
The patent transforms static, pre-set time-based scheduling into dynamic scheduling that adapts to real-time conditions at edge nodes. The system continuously monitors resource availability, service demands, and execution status, then adjusts task scheduling decisions accordingly. This dynamic approach maintains scheduling simplicity through automated decision-making algorithms while significantly improving adaptability to changing real-time conditions.
Solution Approach 2:
The patent enables edge nodes to autonomously make scheduling decisions based on local conditions without requiring complex centralized control. Each edge node independently evaluates its own resource state and service demands, then automatically adjusts task scheduling to optimize local performance. This self-service approach maintains operational simplicity while improving adaptability to real-time conditions through distributed intelligence.
Data Source
AI summary
Disclosed herein are an intelligent scheduling apparatus and method. The intelligent scheduling apparatus includes one or more processors, and an execution memory for storing at least program that is executed by the one or more processors, wherein the at least one program is configured to, in a hybrid cloud environment including a cloud, an edge system, and a near-edge system, configure schedulers for scheduling tasks of the cloud, the edge system, and the near-edge systems, store data, requested by a client, in a work queue by controlling the schedulers based on a scheduler policy and process the tasks based on data stored in the work queue, and collect history data resulting from processing of the tasks depending on the scheduler policy, and train the scheduler policy based on the history data.