Dynamic Data Sharing Method Selection for Distributed Processes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data sharing methods lack flexibility and are often specific to certain resource configurations, limiting their effectiveness in complex computing environments, such as those involving distributed computing and deep learning applications.
Innovation Solution
A method that acquires parameter information for source and target processes, selects a suitable data sharing method based on this information, and implements data sharing across various computing environments, including different processing devices and networks, using techniques like direct memory access and remote direct memory access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed data sharing method is used, then the implementation is simple, but the adaptability to different resource configurations is poor
Solution Approach 1:
The patent implements dynamic data sharing by selecting different sharing methods (memory mapping, message passing, shared memory) based on real-time parameter matching between source and target processes. The system evaluates parameters such as process type, memory address space, and data characteristics to dynamically choose the optimal sharing strategy, transforming a static system into an adaptive one that responds to varying computational environments.
Solution Approach 2:
The patent changes the parameters of the data sharing mechanism by introducing multiple selectable sharing methods with different characteristics. Each method has specific parameter requirements (e.g., memory address compatibility for memory mapping, buffer availability for message passing). The system adjusts which method is active based on whether the current process parameters satisfy the requirements of each sharing method, enabling flexible adaptation without increasing fundamental system complexity.
2Productivity
If distributed computing is implemented, then computing speed is improved, but data sharing efficiency deteriorates
Solution Approach 1:
The patent introduces an intermediary mechanism that automatically selects and manages the appropriate data sharing method based on process parameters. This intermediary layer (the parameter matching and method selection logic) acts as a mediator between distributed processes, choosing the most efficient sharing path without requiring manual configuration or trial-and-error approaches. The intermediary resolves the conflict by intelligently routing data sharing through the most suitable channel for each specific process pair.
Solution Approach 2:
The patent performs preliminary parameter evaluation and method selection before actual data sharing occurs. By pre-assessing whether source and target processes satisfy the requirements of different sharing methods (memory mapping, message passing, etc.), the system prepares the optimal sharing path in advance. This preliminary action prevents wasted time during actual data transfer by ensuring the chosen method is already validated for the current process configuration.
Data Source
AI summary
Embodiments of the present disclosure provide a method, a device, and a computer program product for data sharing. The method includes acquiring first parameter information corresponding to a source process and second parameter information corresponding to a target process, and selecting a desired data sharing method from methods for sharing data between the source process and the target process based on the first parameter information and the second parameter information. The method further includes realizing data sharing between the source process and the target process based on the desired data sharing method. Through this solution, the data sharing efficiency between processes can be improved.