Distributed Processing Computer Management Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional distributed processing techniques require high specifications for management computers to determine which computers execute computing tasks, necessitating efficient scheduling and communication load management.
Innovation Solution
A computer system where processing computers switch between main and management function processing, with a management computer using a schedule table to instruct and manage the execution of management functions, allowing for effective decision-making on processing computer allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the management computer obtains connecting relationship and communication load of all processing computers to decide which computer executes computing, then the processing computer allocation is optimized, but the specification requirement of the management computer becomes high
Solution Approach 1:
The patent divides the management function into two parts: the management computer maintains a simple schedule table showing only the time periods when each processing computer executes management functions, while the processing computers themselves track and report their main function execution status. This segmentation reduces the management computer's burden from managing all scheduling details to merely maintaining time period information.
Solution Approach 2:
The processing computers autonomously determine whether they are executing main functions or management functions based on the schedule table and current time, and self-report their status to the management computer. This self-service mechanism eliminates the need for the management computer to actively query or strictly control each processing computer's state, reducing its computational requirements.
2Measurement precision
If the management computer strictly manages the schedule of processing computers, then the execution timing is controlled precisely, but the resource specification requirement increases
Solution Approach 1:
The schedule table is pre-configured with time periods indicating when each processing computer should execute management functions. The management computer only needs to maintain this predetermined schedule information, while processing computers autonomously determine their current state by comparing the current time with the schedule table, eliminating the need for continuous active management.
Solution Approach 2:
Processing computers automatically report their main function execution status and time to the management computer. This feedback mechanism allows the management computer to maintain precise schedule information with minimal processing, as the processing computers provide real-time status updates without requiring complex management intervention.
Data Source
AI summary
A computer system includes a plurality of processing computers which switches an execution time of a main function processing and a management function processing in a given time, and a management computer which has a schedule table showing a schedule in which each processing computer may execute the management function processing, and which is capable of instructing the execution of the management function processing to each processing computer, wherein the processing computer notifies a start time an end time of the management function processing to the management computer before starting the management function processing, and the management computer updates the schedule table with the notification, an decides the processing computer that may execute the management function processing, by referring to the schedule table, upon instructing the management function processing to the processing computer.