Network Application Orchestration for Edge-Cloud Resource Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high cost and resource inefficiencies in using communication and computing systems for IoT applications, particularly due to the varying costs and performance of public and private networks, lead to potential resource shortages and increased expenses when not managed optimally.
Innovation Solution
A method is implemented to monitor traffic data in networks and migrate applications from private to public networks based on compute resources required per data traffic, selecting applications with lower values of computing resources per data traffic to be migrated first, thereby optimizing resource utilization and reducing costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If applications are migrated from private network to public network to avoid resource shortages, then communication resource availability is improved, but cost increases and performance decreases
Solution Approach 1:
The system changes the parameter of application selection by using compute resources per data traffic ratio as the decision criterion. This parameter change enables differentiated migration strategies where applications are selected based on their resource efficiency rather than uniform migration, thus optimizing the balance between resource availability and cost.
Solution Approach 2:
Instead of migrating all applications when resource shortage occurs, the system applies partial action by selectively migrating only those applications with lower compute resources per data traffic ratios. This partial migration approach prevents unnecessary cost increases while still addressing the resource shortage problem.
2Quantity of substance
If applications are migrated from private network to public network, then communication resource availability is improved, but performance decreases
Solution Approach 1:
The system changes the selection parameter from simple resource availability to compute resources per data traffic ratio. This parameter transformation enables the system to prioritize migrating applications that are less performance-critical, thereby maintaining overall system performance while improving resource availability.
Solution Approach 2:
The system applies different migration priorities to different applications based on their local characteristics (compute resources per data traffic ratio). Applications with lower ratios are migrated first, while those with higher ratios remain on the private network, creating a differentiated quality distribution that preserves overall performance.
3Quantity of substance
If compute resources are allocated to handle increased traffic, then resource availability is improved, but cost increases
Solution Approach 1:
The system changes from allocating compute resources based on traffic volume alone to allocating based on compute resources per data traffic ratio. This parameter change enables more efficient resource allocation where applications with lower ratios receive priority, reducing unnecessary compute resource allocation and associated costs.
Solution Approach 2:
The system enables applications to effectively self-select for migration based on their inherent compute resources per data traffic characteristics. Applications with lower ratios naturally migrate first, creating an automatic resource optimization mechanism that reduces cost without requiring complex external allocation decisions.
Data Source
AI summary
Systems and methods for managing communications between edge devices and cloud devices over a first network (e.g., a private network such as a private 5G network) and a second network (e.g., a public network such as a public 5G network), involving monitoring an amount of traffic data in the first network. When the amount of traffic data exceeds a predetermined value, systems and methods further involve selecting one or more applications executing in a device of the first network; migrating the selected one or more applications from the first network to the second network; and changing a destination address of data communications associated with the selected one or more applications from the device in the first network to another device in the second network; wherein the selecting the one or more applications is conducted based on compute resources required for executing each one or more applications per traffic data.