Distributed Scheduler for Actor Resource Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In communications networks, managing resource units for distributed runtime environments and actors is challenging due to limited resource availability and the complexity of handling multiple devices and services across different network operators, leading to difficulties in efficiently scheduling operations and avoiding service disruptions.
Innovation Solution
A scheduler method that obtains the total available resource units and estimated usage for each instance, scheduling operations to ensure usage remains within available limits, thereby adjusting instance operations to match available resources and avoid communication with a centrally controlled charging system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centrally controlled charging system is used to manage resource units, then resource allocation can be controlled, but system complexity and communication overhead increase
Solution Approach 1:
The patent segments the centralized charging system into distributed scheduler components deployed across multiple runtime environments. Each scheduler instance independently manages resource units for local actor instances, eliminating the need for a single centralized control point and reducing system complexity while maintaining resource allocation control.
Solution Approach 2:
The distributed schedulers enable runtime environments to self-manage their resource unit allocation locally. Each scheduler obtains available resource units and estimated usage for its local instances, then autonomously schedules operations without requiring continuous communication with a central charging system, reducing communication overhead and system complexity.
2Productivity
If resource units are shared across multiple runtime environments and devices, then resource utilization efficiency improves, but managing permissions and avoiding overuse becomes more difficult
Solution Approach 1:
The patent implements feedback mechanisms where each distributed scheduler continuously monitors the usage of resource units by local actor instances against the total available resource units. The scheduler adjusts scheduling decisions based on this feedback to ensure usage remains within limits, enabling efficient resource sharing across multiple runtime environments without requiring complex centralized permission management.
3Adaptability or versatility
If actor instances are distributed across multiple devices and runtime environments, then service scalability improves, but coordinating resource usage and preventing service disruptions becomes more complex
Solution Approach 1:
The patent segments the resource management function into independent distributed scheduler components deployed across multiple runtime environments. Each scheduler manages resource units for local actor instances autonomously, eliminating the need for complex coordination mechanisms while enabling service scalability across multiple devices and network operators.
4Measurement precision
If centralized charging system communication is required for each instance scheduling, then resource allocation accuracy improves, but service continuity and response time deteriorate
Solution Approach 1:
The patent implements preliminary action by having distributed schedulers obtain the total available resource units and estimated usage requirements for each actor instance before scheduling operations. This advance resource assessment enables accurate resource allocation decisions to be made locally without requiring real-time communication with a centralized charging system, ensuring both accuracy and service continuity.
Data Source
Figure 1~16
Figure 2~3
Figure 4~5(f)
AI summary
There is provided mechanisms for scheduling operation of instances of actors on a runtime environment during a time period. A method is performed by a scheduler. The method comprises obtaining a total amount of available resource units for each of the instances to use during the time period. The method comprises obtaining an estimated usage of resource units per instance for the time period. The method comprises scheduling operation of the instances during the time period such that the estimated usage of resource units per instance is within each respective total amount of available resource units.