Redundant Data Storage With Timestamp Certification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprise-class data storage systems face challenges in scalability and high upfront costs due to the need for customized, high-reliability hardware components, and existing solutions for fault tolerance using commodity storage systems require coordinated operation and synchronization of redundant components, which is complex and costly.
Innovation Solution
A system and method for writing and reading redundant data using commodity storage devices, where data is replicated across multiple devices with timestamps and signatures, allowing for decentralized operation and transparent accommodation of failures and malicious attacks, ensuring data consistency and availability through a coordinated certification process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If customized high-reliability hardware components are used, then reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses data replication by storing multiple copies of the same data across different storage devices. Instead of relying on specialized hardware, the system creates redundant copies of data that can be accessed if the primary copy becomes unavailable, thereby achieving high reliability through software-based replication rather than complex customized hardware
Solution Approach 2:
The patent employs commodity storage devices that are inexpensive and easily replaceable rather than expensive customized hardware. The system can tolerate failures of these cheap devices by recovering data from replicas, making the system more cost-effective while maintaining reliability through the replication mechanism
2Reliability
If data replication is used, then fault tolerance is improved, but coordination complexity increases
Solution Approach 1:
The patent implements a self-service mechanism where storage devices autonomously certify data copies using cryptographic signatures and timestamps without requiring complex centralized coordination. Each device independently validates data integrity through digital signatures, eliminating the need for intricate synchronization protocols while maintaining fault tolerance
Solution Approach 2:
The system uses cryptographic feedback mechanisms where each storage device signs data copies with its private key, creating verifiable proofs of authenticity. This feedback loop allows the system to automatically detect and respond to failures or malicious modifications without complex coordination, as any tampering is immediately detectable through signature verification
3Adaptability or versatility
If multiple storage devices are used, then scalability is improved, but management complexity increases
Solution Approach 1:
The patent creates a universal data format that includes embedded cryptographic signatures and timestamps, allowing any storage device in the distributed system to handle any data copy uniformly. This universal approach simplifies management as all devices operate with the same protocols and data structure, enabling easy scalability without increasing management complexity
Data Source
AI summary
In accordance with an embodiment of the invention, a method of writing and reading redundant data is provided. Data is written by storing a copy of the data along with a timestamp and a signature at each of a set of storage devices. The data is read by retrieving the copy of the data, the timestamp and the signature from each of a plurality of the set of data storage devices. One of the copies of the data is selected to be provided to a requestor of the data. Each of the storage devices of the set is requested to certify the selected copy of the data. Provided that a proof of certification of the selected copy of the data is valid, the storage devices of the set are instructed to store the selected copy of the data along with a new timestamp.