Cloud Compute Instance Scheduling for Terminal Task Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Terminal devices with limited computing resources struggle to execute high-computing requirement applications, leading to degraded user experience and limited functionality.
Innovation Solution
A method for scheduling compute instances on a far-side site through a cloud resource scheduling system, allowing terminal devices to offload processing tasks to allocated compute instances, thereby unbinding high-computing requirements from local device performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If terminal devices are configured with high-performance computing resources (CPU, GPU, NPU) to meet increasing computing requirements, then computing capability is improved, but power consumption increases severely
Solution Approach 1:
The patent extracts the high-performance computing resources from the terminal device and relocates them to a remote server. The terminal device only retains the client application and interface, while the computationally intensive tasks are performed on the server's high-performance computing resources. This extraction resolves the contradiction by eliminating the need for terminal devices to have powerful local computing capabilities, thereby reducing power consumption while maintaining computing capability through cloud access.
Solution Approach 2:
The patent introduces a server as an intermediary between the terminal device and the computing tasks. The server acts as a mediator that receives computing requests from the terminal device, executes the tasks using its high-performance computing resources, and returns the results. This intermediary approach allows terminal devices to access powerful computing capabilities without needing to possess them locally, thus resolving the power consumption issue.
2Adaptability or versatility
If terminal devices are configured with high-performance computing resources to run AI, VR/AR, and 3D applications, then application functionality is improved, but device volume increases
Solution Approach 1:
The patent extracts the computing resources required for AI, VR/AR, and 3D applications from the terminal device and places them on a remote server. This allows terminal devices to maintain compact sizes while still accessing high-performance computing capabilities for advanced applications. The device volume is reduced because the heavy computing hardware is no longer needed locally.
Solution Approach 2:
The server provides universal computing resources that can serve multiple types of applications (AI, VR/AR, 3D, etc.) and multiple terminal devices simultaneously. Instead of each terminal device needing its own dedicated high-performance computing hardware for each application type, a single server with universal computing resources serves all these functions for all connected devices, enabling compact terminal designs.
3Productivity
If terminal devices are configured with high-performance computing resources to meet strict requirements on computing capabilities and real-time performance, then application effect is improved, but user experience degrades due to sacrifices and tailoring
Solution Approach 1:
The patent extracts the burden of configuring and managing high-performance computing resources from the terminal device user. Users no longer need to make sacrifices or tailoring decisions about device configuration; they simply access the server's computing resources through a standard client interface. This resolves the user experience degradation by eliminating the need for users to engage in complex configuration decisions.
Solution Approach 2:
The server automatically manages and allocates computing resources based on application requirements and user needs. The system performs self-service by dynamically provisioning computing power without requiring user intervention or configuration decisions. Users simply request computing resources through the client application, and the server automatically provides appropriate resources, improving ease of operation.
4Use of energy by moving object
If terminal devices with low configurations are used, then power consumption and device volume are reduced, but the devices can hardly run applications with high computing requirements
Solution Approach 1:
The server acts as an intermediary that bridges the gap between low-power terminal devices and high-performance computing requirements. The terminal device with low configuration sends computing tasks to the server, which executes them using high-performance resources and returns the results. This intermediary approach allows low-power devices to access high computing capability without needing to possess it locally.
Solution Approach 2:
The terminal device creates a copy or representation of the computing task and sends it to the server for execution. Instead of executing the actual computationally intensive operations locally, the device copies the task definition and parameters to the server, which then performs the actual computation. This copying approach allows low-power devices to leverage remote computing power.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
A method for scheduling a compute instance, an apparatus, and a system are disclosed, and relate to the field of communications technologies. In the method, a client obtains resource requirement information of an application APP running on a terminal device, where the resource requirement information is used to indicate a resource requirement of the APP; the client sends the resource requirement information to a cloud resource scheduling system; the client receives resource allocation information from the cloud resource scheduling system, where the resource allocation information includes information about at least one compute instance allocated by the cloud resource scheduling system based on the resource requirement information, and the at least one compute instance is deployed on at least one far-side site managed by the cloud resource scheduling system; and during running of the APP, the client sends some processing tasks of the APP to the at least one compute instance, and receives results of processing the some processing tasks of the APP by the at least one compute instance. This unbinds a service with a high computing requirement from computing performance of the terminal device, thereby providing unlimited possibilities for a computing terminal.