Dynamic Edge and Local Accelerator Resource Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In compute devices, certain functions require acceleration that general-purpose processors cannot provide efficiently, especially when battery power is depleted, making it infeasible to utilize existing accelerator devices due to power inefficiency.
Innovation Solution
A system for dynamic selection of edge and local accelerator resources, which includes a client compute device communicating with an edge gateway to query available accelerator resources, comparing their properties with local resources and selecting the best resources based on acceleration selection factors such as latency, power usage, and security features to execute functions efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If accelerator devices are used to execute functions faster, then execution speed is improved, but power consumption increases
Solution Approach 1:
The system dynamically selects between local and edge accelerator resources based on real-time conditions such as battery power levels, network availability, and computational requirements. This dynamic adaptation allows the system to use high-performance local accelerators when power is abundant and switch to edge accelerators when power is constrained, resolving the contradiction between execution speed and power consumption
Solution Approach 2:
The edge gateway acts as an intermediary between the client device and edge accelerator resources. It manages the coordination and resource allocation, enabling the system to leverage external accelerator resources when appropriate while maintaining local execution capability when needed, thus balancing speed and power consumption
2Speed
If local accelerator resources are used, then execution speed is improved, but adaptability to changing power conditions deteriorates
Solution Approach 1:
The system implements dynamic resource selection that adapts to changing power conditions by evaluating battery status, network state, and application requirements in real-time. This allows the system to maintain high execution speed through local accelerators when power is available while automatically adapting to use edge resources when power conditions change, thus maintaining both speed and adaptability
3Use of energy by moving object
If edge accelerator resources are used, then power consumption is reduced, but latency increases
Solution Approach 1:
The system dynamically evaluates network conditions and computational urgency to determine whether to use edge or local accelerators. For time-critical functions, it prioritizes local execution despite higher power consumption, while for less time-sensitive tasks, it uses edge resources to reduce power consumption, thus balancing latency and power consumption based on real-time conditions
Solution Approach 2:
The system changes operational parameters such as accelerator selection, power modes, and execution timing based on evaluated conditions. By adjusting these parameters dynamically, the system can optimize the trade-off between latency and power consumption for different execution scenarios
Data Source
AI summary
Technologies for providing dynamic selection of edge and local accelerator resources includes a device having circuitry to identify a function of an application to be accelerated, determine one or more properties of an accelerator resource available at the edge of a network where the device is located, and determine one or more properties of an accelerator resource available in the device. Additionally, the circuitry is to determine a set of acceleration selection factors associated with the function, wherein the acceleration factors are indicative of one or more objectives to be satisfied in the acceleration of the function. Further, the circuitry is to select, as a function of the one or more properties of the accelerator resource available at the edge, the one or more properties of the accelerator resource available in the device, and the acceleration selection factors, one or more of the accelerator resources to accelerate the function.