Multi-Cloud Coded Block Storage for Secure High Availability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud-based block storage systems face challenges in availability and security, with limited ability to block access once compromised and lower availability compared to on-premises storage, and techniques like local mirroring are not feasible due to dependency between disks.
Innovation Solution
A multi cloud-based storage system generates and distributes coded blocks across multiple clouds, using an error-correcting code like Reed-Solomon, allowing data recovery from any k out of n clouds and blocking access from compromised clouds, ensuring high availability and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data is stored in cloud-based block storage, then storage capacity and accessibility are improved, but security control and availability are worsened compared to on-premises storage
Solution Approach 1:
The patent segments data into multiple coded blocks and distributes them across different cloud storage locations. This segmentation allows the system to maintain security control by requiring multiple blocks for data reconstruction while preserving cloud storage accessibility benefits
Solution Approach 2:
The patent introduces an intermediary encoding layer (error-correcting codes) between the original data and cloud storage. This intermediary mechanism enables security control and availability management without directly compromising cloud storage accessibility
2Reliability
If local mirroring is used for data redundancy, then data availability is improved, but device dependency and complexity increase
Solution Approach 1:
Instead of mirroring entire data sets across dependent disks, the patent segments data into coded blocks that can be independently stored across different cloud locations, reducing device dependency while maintaining availability
Solution Approach 2:
The patent changes the redundancy parameter from exact copies (mirroring) to coded representations, allowing data reconstruction from multiple different block combinations without requiring identical disk configurations
3Reliability
If coded blocks are distributed across multiple clouds, then security and availability are improved, but storage complexity and latency increase
Solution Approach 1:
The patent employs universal error-correcting code mechanisms that can operate across multiple cloud platforms with different characteristics, managing complexity through a unified approach while maintaining security and availability benefits
Data Source
AI summary
A method, apparatus, and system for storing data at a multi cloud-based storage system is disclosed. The operations comprise: receiving a data block at a first cloud of a multi cloud-based storage system for storage, the multi cloud-based storage system comprising a first number (n) of clouds; generating the first number (n) of coded blocks at the first cloud based on the data block, wherein the data block is recoverable from any second number (k) out of the first number (n) of coded blocks, and wherein the second number (k) is greater than 1 and less than the first number (n); and distributing, by the first cloud, the first number (n) of coded blocks to the first number (n) of clouds of the multi cloud-based storage system, each of the clouds including the first cloud receiving a respective one of the first number (n) of coded blocks for storage.