Distributed Computing via Embedded Processor Network Mediation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computing systems require significant resources to process large datasets, such as those from the SETI project, and are inefficient in utilizing idle processing power in devices like mobile phones and other embedded processors.
Innovation Solution
Establishing a network of embedded processors in various devices to distribute computing tasks during idle times, partitioning problems into segments based on processor capabilities and responsibilities, and aggregating partial solutions to solve complex problems efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed computing is implemented using embedded processors in portable devices, then processing capacity and productivity are improved, but device complexity and network management complexity increase
Solution Approach 1:
A server acts as an intermediary between the distributed embedded processors and the problem data. The server receives the problem, partitions it into segments, distributes segments to appropriate processors based on their capabilities, collects partial solutions, and synthesizes the final result. This mediator approach manages the complexity of coordinating multiple distributed processors while enabling enhanced processing capacity.
Solution Approach 2:
The problem is divided into multiple smaller segments that can be processed independently by different embedded processors. Each processor works on its assigned segment and returns a partial solution. This segmentation enables parallel processing across multiple devices, significantly improving overall productivity while allowing each processor to handle manageable task complexity.
2Productivity
If problem segments are distributed to multiple processors based on processor classes, then processing efficiency is improved, but communication overhead and system complexity increase
Solution Approach 1:
The system assigns different problem segments to processors based on their local qualities - specifically their processor class, computational capabilities, and current availability. The server evaluates each processor's characteristics and matches segments accordingly, ensuring that each processor handles tasks appropriate to its capabilities. This local quality matching improves processing efficiency while avoiding the need for complex universal scheduling algorithms.
Solution Approach 2:
The system uses processor class as a key parameter for distributing problem segments. By categorizing processors into different classes based on their capabilities, the system simplifies the distribution decision-making process. This parameter-based approach enables efficient load balancing and task allocation without requiring complex real-time negotiations between processors.
3Productivity
If embedded processors are utilized for distributed computing, then idle computing resources are effectively used, but reliability and data security challenges arise
Solution Approach 1:
The server serves as a trusted intermediary that manages data flow between the external problem source and the distributed embedded processors. It receives the original problem, partitions it into segments, distributes them securely to processors, collects partial solutions, and synthesizes the final answer. This intermediary approach enables effective utilization of idle processor resources while maintaining data security through centralized control and verification.
Data Source
AI summary
A method of distributed computing is provided. The method comprises establishing a network of processors, wherein the processors are associated with a plurality of classes and wherein at least some of the processors are embedded in portable electronic devices, and partitioning a problem to be solved by the network of processors into a plurality of problem segments, wherein the problem segments are associated with the plurality of classes associated with the processors. The method further comprises transmitting the problem segments to the processors based on the class of each processor, receiving partial solutions from the processors, and analyzing the partial solutions. An aspect of the method may comprise defining the classes based on at least one of a processor class, a data communication quality of service, and a responsibility level of the processor.