Asynchronous Data Dispersal for Disaster Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current datacenter disaster recovery systems face challenges in efficiently replicating data across multiple storage sites, leading to high costs and bandwidth consumption, while existing solutions do not effectively address the need for asynchronous data dispersal and dynamic storage system behavior.
Innovation Solution
A method for configuring an asynchronous Disaster Recovery (DR) process that calculates information dispersal parameters to disperse slices of DR enabling data between storage sites, considering Service Level Specifications (SLS) and dynamic behavior, allowing for efficient data reconstruction even if a number of storage sites fail, and includes features like recalculating parameters upon SLS breaches and utilizing additional resources as needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is replicated to multiple DR storage sites, then data availability and reliability are improved, but storage space consumption and bandwidth usage increase
Solution Approach 1:
The patent divides the DR enabling data into multiple slices and disperses them across different storage sites. Instead of replicating complete data copies to each DR site, the system segments the data into portions that can be reconstructed together, reducing the storage space required at each individual site while maintaining overall data availability.
Solution Approach 2:
The system changes the parameter of data representation by using erasure coding to transform original data into DR enabling data with different properties. This transformation allows the same level of reliability to be achieved with reduced storage overhead, as the encoded data can be reconstructed from fewer than the total number of slices stored across distributed sites.
2Reliability
If data is replicated to multiple DR storage sites, then data availability is improved, but bandwidth consumption increases
Solution Approach 1:
By segmenting data into slices and using erasure coding, the system reduces the total amount of data that needs to be transmitted to DR sites. Only the encoded slices are sent to remote locations, rather than complete data copies, thereby reducing bandwidth consumption while maintaining the ability to reconstruct original data from the distributed slices.
3Reliability
If synchronous replication is used with remote storage sites, then data consistency is improved, but communication latency increases
Solution Approach 1:
Instead of using synchronous replication that waits for confirmation from remote sites, the system inverts the approach by using asynchronous dispersal of encoded slices. The primary site can continue operating without waiting for remote site confirmations, and data consistency is maintained through the mathematical properties of erasure coding rather than through synchronous acknowledgment protocols.
Data Source
AI summary
Method and system for asynchronously dispersing Disaster Recovery (DR) enabling data between a plurality of storage sites. The method comprises: receiving, at a primary storage site, a written block and a write frequency counter associated with the written block. In case the write frequency counter is below a threshold: receiving information dispersal parameters including number indicative of a size difference between said written block and DR enabling data based on said written block; number of slices to slice said DR enabling data into and data indicative of DR storage sites of said plurality of storage sites for storing said slices. Further calculating DR enabling data based on written block, wherein DR enabling data is larger than said written block by size difference; slicing DR enabling data in accordance with number of slices; and dispersing slices in accordance with data indicative of DR storage sites.