ML-Based Resource Distribution for Edge Device Task Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computing devices often lack sufficient resources to execute tasks, leading to delayed or cancelled operations, which can have detrimental consequences.
Innovation Solution
A machine-learning-based digital communication system that includes a central server and edge devices, where the ML engine calculates predicted resource availability scores to distribute tasks among edge devices, optimizing resource utilization and ensuring tasks are executed efficiently by devices with sufficient resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a computing device is directed to execute a computing task, then the task execution is initiated, but the device may have insufficient resources leading to delayed or cancelled execution
Solution Approach 1:
The patent combines multiple edge devices into a unified computing resource pool managed by a central server. Instead of relying on a single device's resources, the system merges resources across multiple devices to ensure task execution, resolving the contradiction between initiating tasks and having sufficient resources available.
Solution Approach 2:
The system dynamically assigns computing tasks to edge devices based on real-time resource availability conditions. The central server monitors and adjusts task distribution dynamically, allowing the system to adapt to changing resource states and ensure tasks are executed by devices with adequate resources.
2Productivity
If tasks are distributed to edge devices based on resource availability, then resource utilization is maximized, but system complexity increases due to ML engine and coordination requirements
Solution Approach 1:
The central server acts as an intermediary between the ML engine and edge devices, managing the complexity of task distribution and resource coordination. This intermediary structure allows the system to maximize resource utilization through intelligent algorithms while shielding individual edge devices from complex coordination requirements.
Solution Approach 2:
Each edge device independently reports its own resource availability status to the central server. This self-service approach allows devices to provide accurate real-time information about their capacity without requiring complex centralized monitoring of each device's internal state, simplifying the overall system architecture.
3Reliability
If computing tasks are delayed or cancelled due to insufficient resources, then resource constraints are respected, but detrimental consequences occur from task non-execution
Solution Approach 1:
The patent transitions from a single-dimension resource constraint model (one device at a time) to a multi-dimensional resource pool across multiple devices. This dimensional change allows the system to satisfy resource constraints while avoiding task delays by finding available resources in other devices when the originally assigned device is unavailable.
Data Source
AI summary
Systems and methods for maximizing resource utilization in a digital communication system are provided. A method may include receiving a direction for a first one of a plurality of edge devices to execute a task, wherein the edge devices communicate with each other and with a central server via a communication network. The central server may be operated by an entity that is independent of the communication network. The plurality of edge devices may each include an authenticated software application that is provided by the entity. The method may also include calculating, via the ML engine for each of the plurality of edge devices, a predicted resource availability score, and distributing, via the central server, the task among the plurality of edge devices based on the predicted resource availability scores.