Parity-Protected Aggregates for Efficient Data Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face inefficiencies and high costs in maintaining redundancy for data accessibility, particularly when aggregate failures occur, as traditional mirroring methods consume excessive resources and result in slow response times.
Innovation Solution
The implementation of redundancy-protected aggregates across a cluster of interconnected nodes, where data is striped across multiple aggregates, with each aggregate acting as a parity owner for specific regions, allowing for efficient reconstruction of data in case of failures and balanced resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional mirroring methods are used to protect against aggregate failure, then data accessibility is maintained, but storage space efficiency deteriorates (utilizing half of the overall space consumed on a storage system)
Solution Approach 1:
The patent uses parity copies instead of complete data mirrors. By storing redundancy information (parity) rather than full duplicate data, the system maintains data accessibility while significantly reducing the storage overhead from 50% to approximately 33% or less depending on the RAID configuration.
Solution Approach 2:
The patent changes the redundancy parameter from full mirroring (100% duplicate) to parity-based redundancy (fractional duplicate). This parameter change allows the system to maintain the same reliability level while improving storage efficiency by storing only the differential information needed for reconstruction.
2Reliability
If mirrored synchronous storage systems are used to protect against aggregate failure, then data accessibility is maintained, but response time deteriorates (especially slow because such systems store data in both mirrors before responding)
Solution Approach 1:
The patent implements partial redundancy through parity rather than complete mirroring. The system stores data locally and maintains only the minimum necessary redundancy information, allowing faster write operations since the system doesn't need to complete full synchronization with remote mirrors before responding to clients.
Solution Approach 2:
The patent segments the storage system into local aggregates with distributed parity information across the cluster. This segmentation allows the system to respond to read requests from the local aggregate immediately, while parity updates are handled asynchronously, improving response time compared to synchronous mirroring.
3Reliability
If complete duplication of storage system resources is used to create mirrored images, then aggregate failure protection is achieved, but operating costs deteriorate (substantial operating costs)
Solution Approach 1:
The patent makes storage resources universal by allowing any aggregate in the cluster to serve as parity for any other aggregate. This multi-functionality eliminates the need for dedicated mirror resources, as each storage resource can dynamically serve multiple redundancy roles across different aggregates, reducing overall resource requirements and operating costs.
Data Source
AI summary
The present invention provides a system and a method for utilizing a parity protection module to back up data on striped aggregates. Specifically, the system computes parity data for data stored at a particular location of each of a plurality of constituent aggregates, and stores the parity on one of the constituent aggregates that is a parity owner for that particular location of data. In the event one of the constituent aggregates fails, new data may still be accessed by the system (the striped aggregates), both to write new data, and to read data stored on the failed aggregate. In particular, the parity protection module allows clients to read data from a failed aggregate by running a reverse parity computation, which may also be used to restore the data to the failed aggregate.