Distributed Computing Task Costing for Mobile Device Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current distributed computing systems using mobile devices face inefficiencies due to high bandwidth consumption and energy costs, as they rely heavily on remote data processing, leading to network scalability issues and device energy drain, especially when handling computationally intensive tasks.
Innovation Solution
Implementing a distributed computing system that allows mobile devices to process tasks locally, utilizing nearby devices as a cluster to share computational loads, thereby reducing the need for remote data transfer and leveraging local high-bandwidth Wi-Fi networks, and opportunistically using cellular networks during off-peak hours for data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If mobile devices rely heavily on remote data processing through cellular networks, then computational tasks can be performed, but bandwidth consumption and energy costs increase significantly
Solution Approach 1:
The patent segments the distributed computing task into multiple subtasks that are distributed across multiple mobile devices in the cluster. Each device processes a portion of the overall task locally, reducing the need for continuous remote communication and thereby lowering energy consumption while maintaining productivity.
Solution Approach 2:
The patent introduces a local cluster of mobile devices as an intermediary between the thin client and the back-end computing system. This intermediary cluster handles computationally intensive tasks locally through peer-to-peer communication, reducing reliance on cellular network bandwidth and decreasing energy costs associated with remote data processing.
2Productivity
If mobile devices use cellular networks for data transfer, then remote processing can be accessed, but network scalability issues and congestion occur
Solution Approach 1:
The patent transitions from a centralized remote processing model to a distributed local processing model by introducing a cluster of mobile devices. This dimensional change in system architecture allows tasks to be processed across multiple local nodes rather than through a single cellular network path, reducing bandwidth consumption and improving network scalability.
3Power
If computationally intensive tasks are processed remotely, then processing power is sufficient, but device energy drain increases
Solution Approach 1:
The patent merges the computational resources of multiple mobile devices in proximity to form a distributed cluster. By combining processing power across multiple devices, the system can handle computationally intensive tasks locally without requiring any single device to consume excessive energy, thereby reducing overall energy drain while maintaining sufficient processing power.
Data Source
AI summary
Distributed computing task costing is disclosed. Costing can be employed to determine if a task will be passed to a distributed computing cluster including mobile devices. Costing can include determining a base cost value predicated on a selectable level of utility to a user of a burdened device, a base time value related to completing the task without the use of the cluster, determining a delay cost, and any monetary costs associated with performing the task without the use of the cluster. Costing can further include demining a relief cost that can include the selectable level of utility, a relief time value related to completing the task with the cluster, the delay cost, an incentive cost based on the sum across a set of relief devices and their corresponding parameters for cluster participation, and any remaining monetary costs borne by the burdened device. Where the base cost value and relief cost value satisfies conditions, a task can be divided into subtasks that can be distributed to the cluster to accomplish the task in a distributed computing environment.