Parallel Processing Controller Command Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In high-density computer systems, the increased number of components due to high-density mounting leads to a higher system control load, resulting in longer system control times and reduced operational simplicity, as existing control methods require duplicate reporting of infrequently changing information.
Innovation Solution
A parallel processing device with a management unit and controller that includes a command storage, type identification unit, and generator, which interpolates omitted control information from previously stored commands to reduce duplicate reporting and communication overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If high-density mounting is implemented to reduce installation area cost, then the number of components per system increases, but the system control load on the service processor increases significantly
Solution Approach 1:
The patent divides the control architecture into multiple distributed controllers, each responsible for specific boards or components. Instead of one service processor handling all control tasks for the entire system, control functions are segmented and distributed to individual controllers that can independently manage their respective components, thereby reducing the control load on any single controller.
Solution Approach 2:
The patent introduces a hierarchical control dimension by adding intermediate controllers between the service processor and individual components. This creates a multi-level control structure where the service processor manages high-level system coordination while intermediate controllers handle detailed component control, effectively distributing the control burden across different hierarchical levels.
2Adaptability or versatility
If the number of components to be controlled increases, then system functionality improves, but the control time and processing time for system operations increase
Solution Approach 1:
The patent implements preliminary action by having controllers pre-load and cache frequently accessed control data and configuration information in local memory. This allows controllers to quickly retrieve necessary information without repeatedly querying the service processor, significantly reducing control time for routine operations while maintaining full system functionality.
Solution Approach 2:
The patent creates local copies of control data and configuration information in the memory of individual controllers. Instead of the service processor storing and managing all control data centrally, each controller maintains local copies of the data it needs, enabling parallel access and eliminating bottlenecks in data retrieval during system operations.
3Measurement precision
If complete command information is transmitted from management unit to controller, then control accuracy is maintained, but communication overhead increases due to duplicate reporting of infrequently changing information
Solution Approach 1:
The patent extracts and separates frequently changing control parameters from static or infrequently changing configuration data. The command protocol is designed to transmit only the necessary dynamic parameters in each communication cycle, while static parameters are transmitted less frequently or cached locally. This extraction of essential information reduces communication overhead while maintaining control accuracy for time-critical parameters.
Solution Approach 2:
The patent implements periodic action by transmitting different types of command information at different intervals. Time-critical control commands are transmitted with high frequency, while configuration and status information that changes slowly are transmitted at lower frequencies. This periodic transmission strategy reduces overall communication overhead while ensuring that critical control signals maintain their required update rates for accurate system control.
Data Source
AI summary
A parallel processing device includes a management unit, a plurality of nodes, and a controller that controls each of the plurality of nodes in accordance with a first command transmitted from the management unit. The controller includes a command storage that stores a second command generated a previous time, a command type identification unit that identifies a command type of the first command transmitted from the management unit, and a command generator that generates a third command by using the second command according to the command type.