Encrypted Index Files for Secure Cloud Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing solutions for secure data storage in cloud environments face challenges in searching encrypted data without compromising security, especially when network connectivity is unavailable, leading to downtime and disruption in data uploading and retrieval processes.
Innovation Solution
A method and system for secure data storage and retrieval that involves creating and encrypting index information locally with a tenant private key, uploading encrypted index files to the cloud, and querying and decrypting them efficiently to facilitate seamless data retrieval without network connectivity, using chronological indexing and compression to optimize performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is encrypted before storage in cloud with private keys, then security of PHI data is improved, but searching capability of encrypted data deteriorates
Solution Approach 1:
The patent segments the data into two parts: encrypted PHI data stored in cloud buckets and separate encrypted index files stored in cloud tables. The index files contain searchable keywords and their corresponding encrypted bucket locations, enabling search without decrypting the actual PHI data. This segmentation resolves the contradiction by maintaining encryption security while providing search capability through the separate index structure.
2Ease of operation
If complete encrypted data is downloaded to client for decryption and searching, then searching capability is improved, but time consumption and network bandwidth deteriorate
Solution Approach 1:
The patent extracts only the necessary index information (keywords and encrypted bucket location mappings) from the encrypted data and stores it separately in cloud tables. When a search is performed, only these compact index files are downloaded and decrypted, not the entire PHI data set. This extraction principle dramatically reduces download time and network bandwidth consumption while maintaining full search capability.
3Reliability
If data upload is deferred and cached during network unavailability, then security of data transmission is improved, but system responsiveness and data freshness deteriorate
Solution Approach 1:
The patent performs preliminary actions by creating and encrypting index files locally on the client side before network upload. These index files are prepared in advance and can be immediately used for local searches even when network connection is unavailable. When the network becomes available, the pre-prepared index files are uploaded to the cloud, ensuring both security and system responsiveness without requiring real-time network connectivity for index operations.
4Speed
If hierarchical index structures are used for large data sets, then search efficiency is improved, but index update complexity and performance deteriorate
Solution Approach 1:
The patent segments the index into flat, non-hierarchical key-value pairs stored in cloud tables, where each row contains a keyword and its corresponding encrypted bucket location. This flat structure eliminates the complexity of hierarchical index updates and merging operations. The segmentation of index data into manageable rows in a table structure simplifies both reading and updating operations compared to hierarchical structures, while still enabling efficient search through direct key lookups.
Data Source
AI summary
A method for secure storage of data and retrieval of desired data from a cloud-based service environment includes receiving the data from a tenant having a unique tenant ID, dynamically extracting data to be indexed from the received data, and creating index information from the extracted data. The index information is encrypted with a tenant private key, the encrypted index information is uploaded into the cloud environment in the form of index files, and a last uploaded index file is queried for in the cloud environment. The encrypted index information of the index file is decrypted with the tenant private key, the decrypted index information is searched for a relevant patient record, and the corresponding desired data is retrieved from the cloud environment. The desired data is rendered onto a client application. The index files are created from the index information in chronological order of receipt of the data.