Synthetic Data Frameset Store for Distributed Dataset Consistency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for obtaining training datasets for machine learning models are inefficient, expensive, and lack comprehensive functionality, making high-quality datasets inaccessible or too costly for widespread use across different domains.
Innovation Solution
A distributed computing system provides synthetic data as a service (SDaaS) that automates the generation and refinement of training datasets using a service-oriented architecture, incorporating engines like asset assembly, scene assembly, frameset assembly, and crowdsourcing to create programmable machine-learning data representations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is stored in a distributed computing system, then data accessibility and processing capability are improved, but data consistency and reliability deteriorate due to multiple copies and potential node failures
Solution Approach 1:
The patent creates a synthetic copy of the data store frame set that mirrors the actual data store frame set. This synthetic copy includes replicated data frames and metadata, allowing the system to maintain data consistency across multiple nodes while enabling distributed processing. The synthetic copy acts as a virtual representation that synchronizes with the actual data store, resolving the contradiction between distributed accessibility and data consistency.
2Reliability
If a data store frame set is replicated across multiple nodes, then system availability is improved, but synchronization complexity and overhead increase
Solution Approach 1:
The patent segments the data store into discrete data frames that can be independently replicated and synchronized across multiple nodes. Each data frame is a self-contained unit with associated metadata, allowing selective replication only of changed frames rather than the entire data store. This segmentation reduces synchronization overhead while maintaining availability across distributed nodes.
Solution Approach 2:
The patent introduces a synthetic data store frame set as an intermediary layer between the actual data store and the distributed nodes. This synthetic copy acts as a mediator that abstracts the complexity of direct synchronization, providing a unified view of data across nodes while simplifying the synchronization process through replicated metadata and frame references.
3Adaptability or versatility
If data is accessed through multiple distributed nodes, then system scalability is improved, but data access latency and inconsistency increase
Solution Approach 1:
The patent performs preliminary actions by pre-replicating data frames and metadata to multiple nodes before actual data access is needed. The synthetic data store frame set is populated with advance knowledge of data locations and structures, allowing nodes to serve read requests from local copies rather than requiring real-time synchronization. This reduces access latency while maintaining scalability through the distributed frame set architecture.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
Various embodiments, methods and systems for implementing a distributed computing system frameset package store are provided. Initially, a query for a frameset package is received from a frameset package query interface. The frameset query interface comprises a plurality of frameset package categories. A query result frameset package is identified based on a frameset package profile. The query result frameset package is communicated.