AI Task Distribution Across Nearby Devices to Reduce Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current device-cloud based AI collaborative computing solutions experience significant forwarding latency due to the need to send tasks and execution results across multiple networks between terminal devices and cloud data centers.
Innovation Solution
A task processing method where a first device determines if its available computing resources are insufficient for an AI task group and selects nearby collaborative computing devices within a valid communication distance to distribute and execute the tasks, thereby reducing latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If device-cloud based AI collaborative computing is used to execute AI tasks, then computing capability is extended, but forwarding latency increases due to multiple network transmissions
Solution Approach 1:
The patent segments the AI task execution into two parts: local execution on terminal devices and distributed execution on nearby devices within the same area. This segmentation allows critical AI tasks to be processed locally without requiring full cloud communication, reducing forwarding latency while maintaining extended computing capability through selective task distribution to nearby devices.
Solution Approach 2:
The patent introduces nearby devices within the same area as intermediary computing nodes between the terminal and the cloud. These intermediary devices can execute AI tasks locally, acting as a buffer that reduces the need for frequent cloud communications, thereby extending computing capability while minimizing forwarding latency.
2Power
If tasks are sent to cloud data center for execution, then computing power is sufficient, but communication cost and latency increase
Solution Approach 1:
The patent applies local quality by enabling terminal devices and nearby devices to execute AI tasks locally without requiring cloud communication. This local execution capability provides sufficient computing power for many AI tasks while eliminating the communication latency associated with cloud data center interactions.
Solution Approach 2:
The patent adds a new dimension to AI task execution by introducing nearby devices within the same area as an intermediate layer between terminal and cloud. This dimensional addition creates a hierarchical execution model where tasks can be processed at multiple levels (terminal, nearby devices, cloud), reducing dependency on cloud communication while maintaining adequate computing power.
3Productivity
If AI tasks are distributed to multiple devices for collaborative execution, then processing efficiency improves, but system complexity increases
Solution Approach 1:
The patent segments the collaborative computing system into terminal devices and nearby devices within the same area, creating a simplified hierarchical structure. This segmentation improves processing efficiency by distributing AI tasks across multiple devices while reducing system complexity through clear role definition and localized collaboration.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
This application discloses a task processing method and apparatus. The method includes: when a first device determines that an available computing resource is less than a computing resource required by a to-be-executed AI task group, selecting at least one second device from a valid device in a first area, where the first area is an area that uses the first device as a center and a valid communication distance of the first device as a radius, and the valid device in the first area is a collaborative computing device that is in the first area and that is connected to the first device; sending, by the first device, a task in the to-be-executed AI task group to each of the at least one second device, where if the to-be-executed AI task group includes at least two AI tasks, the sent task is some of the at least two AI tasks, or if the to-be-executed AI task group includes one AI task, the sent task is some subtasks of the one AI task; and receiving, by the first device, an execution result of the sent task from the second device.