Cloud Work Distribution Model for Mobile Application Offloading
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile devices face challenges in executing computationally expensive applications due to limited processing power and memory resources, necessitating a method to optimize power consumption and distribute workload effectively between the device and cloud-based resources.
Innovation Solution
A system and method that provisions a computing application on a cloud infrastructure, using a work distribution model to determine which application components should be executed on the device or the cloud, optimizing CPU cycles and minimizing latency and power consumption by delegating processing tasks dynamically based on client device capabilities and network conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If computationally expensive applications are executed on mobile devices, then application functionality is improved, but power consumption increases and battery life decreases
Solution Approach 1:
The application is divided into multiple components or tasks that can be executed separately. Some components are executed on the mobile device while others are executed on remote cloud-based virtual machines, allowing the system to segment computational workload between device and cloud resources to reduce local power consumption.
Solution Approach 2:
A cloud-based virtual machine acts as an intermediary between the mobile device and the application logic. The virtual machine receives requests from the mobile device, executes computationally intensive tasks, and returns results, thereby mediating the computational workload and reducing the energy burden on the mobile device.
2Productivity
If computationally expensive applications are executed on mobile devices, then application functionality is improved, but processing speed may be insufficient due to limited device resources
Solution Approach 1:
The application workload is segmented into compute-intensive tasks and lightweight tasks. Compute-intensive tasks are offloaded to cloud-based virtual machines with superior processing capabilities, while lightweight tasks remain on the mobile device, thereby improving overall processing speed for the application.
Solution Approach 2:
The application state and data are copied between the mobile device and cloud-based virtual machines. The virtual machine maintains a copy of the application state, allowing it to execute computationally expensive operations remotely and return results to the device, effectively bypassing the device's processing limitations.
3Use of energy by moving object
If application components are distributed between device and cloud, then power consumption is reduced, but system complexity increases
Solution Approach 1:
The system automatically manages the distribution of computational tasks between the mobile device and cloud-based virtual machines through a work distribution model. The virtual machine and device autonomously determine which tasks to execute based on predefined criteria, reducing the need for complex manual configuration and management overhead.
Solution Approach 2:
The cloud-based virtual machine serves multiple functions: it executes computationally intensive application logic, maintains application state, handles data processing, and communicates with the mobile device. This multi-functionality consolidates complex operations into a single universal component, reducing overall system complexity despite the distributed architecture.
Data Source
AI summary
A system is described to determine a work distribution model for a computing application deployed on a cloud. In one embodiment, the system comprises a launch request detector, a provisioning module, and a work distribution module. The launch request detector may be configured to detect a request from a client device to launch a computing application. The provisioning module may be configured to provision an application container on a virtual instance of a machine for executing the computing application. The work distribution module may be configured to generate a work distribution model for the computing application, the work distribution model indicating a first portion of the computing application to be executed on the client device and a second portion of the computing application to be executed on the virtual instance of a machine. The client device may be a desktop device or a client device.