Federated Computing Parallelization via Data Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current federated computing methods face challenges in improving performance, particularly in large-scale data privacy computing, where efficiency and accuracy are compromised due to limitations in data processing and parallelization.
Innovation Solution
The method involves obtaining metadata from participants based on a task ID, dividing it into data slices, generating data sets, and processing these sets using corresponding data processing services to achieve parallelization and enhance computing speed through data parallelism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional federated computing methods are used to process large-scale data, then data privacy is protected, but computing speed and efficiency deteriorate
Solution Approach 1:
The patent divides the metadata into n data slices and generates n corresponding data sets, which are then processed by n different data processing services in parallel. This segmentation enables concurrent processing of encrypted data, significantly improving computing speed while maintaining data privacy protection through the federated computing framework.
2Productivity
If data is divided into slices for parallel processing, then computing efficiency improves, but system complexity increases
Solution Approach 1:
The patent employs a unified federated computing platform that can handle multiple data slices and various data processing services through a common architecture. The platform uses standardized interfaces and protocols for data slice management, encryption, and parallel processing, which reduces the operational complexity despite the increased number of processing components.
Solution Approach 2:
The patent introduces a federated computing platform as an intermediary layer that coordinates between the data slices, encryption mechanisms, and processing services. This intermediary manages the complexity of parallel processing by providing centralized control for task distribution, result aggregation, and security management, thereby simplifying the overall system architecture.
3Productivity
If multiple data processing services are called for parallel processing, then computing speed improves, but resource consumption increases
Solution Approach 1:
By dividing the data into n slices and processing them in parallel through multiple services, the system completes the overall computation faster than sequential processing would allow. The parallel execution reduces the total time resources are occupied, thereby improving resource utilization efficiency despite increased concurrent resource demands.
Solution Approach 2:
The patent maintains continuous useful action by keeping multiple data processing services running concurrently on different data slices. This parallel execution ensures that computational resources are continuously utilized across multiple tasks simultaneously, maximizing resource productivity and reducing idle time compared to sequential processing approaches.
Data Source
AI summary
A federated computing method, an electronic device and a storage medium. Metadata to be processed is obtained from each participant based on a task identifier (ID) to be executed. n data slices corresponding to each participant are obtained by dividing the metadata, where n is a positive integer greater than 1. n data sets are generated based on the n data slices corresponding to each participant. n data processing services corresponding to the task ID are called to process each data set using each of the data processing services. A federated computing result corresponding to the task ID is determined based on a processing result of each of the data processing services.