Backup-Based RAG Vector Generation with Sensitive-Data Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing retrieval augmented generation (RAG) systems rely on live enterprise data, which requires significant IT investment and can disrupt the host system, posing potential risks and complications.
Innovation Solution
Utilize backup data managed by a data management system (DMS) to generate vectors based on extracted data, storing them in a vector database accessible to communication applications, thereby avoiding disruptions and enabling streamlined and customizable implementations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If live enterprise data is used for RAG systems, then the accuracy and relevance of LLM responses is improved, but the host system experiences disruption and requires significant IT investment
Solution Approach 1:
The patent creates a copy of enterprise data from backup systems to build the vector database for RAG operations. This copy is made without disrupting the host system's live operations, allowing accurate responses to be generated while avoiding the harmful effects of accessing production systems directly.
Solution Approach 2:
The patent introduces a data management system as an intermediary layer between the host system and the RAG application. This intermediary accesses backup data rather than live data, mediating the interaction to prevent system disruption while still providing accurate enterprise information to the LLM.
2Reliability
If backup data is used instead of live data, then system disruption is avoided and IT investment is reduced, but data freshness and relevance may be compromised
Solution Approach 1:
The system performs preliminary actions by maintaining backup data copies in advance of when they are needed for RAG operations. This allows the vector database to be populated with enterprise data beforehand without disrupting live systems, ensuring both system stability and data availability when queries are made.
Data Source
AI summary
Methods, apparatuses, and computer readable media are configured to perform operations comprising: obtaining, by a data management system (DMS), a first snapshot of a computing system, wherein the first snapshot comprises data associated with a set of files; determining, by the DMS, from among the set of files, a first subset of files or portions of files that comprise sensitive information; generating, by the DMS, one or more vectors based at least in part on data associated with a second subset of files or portions of files from among the set of files, the second subset of files or portions of files exclusive of files from the first subset of files or portions of files; and adding, by the DMS, the one or more vectors to a vector database along with metadata or a pointer to the metadata.


