Asynchronous Recovery of Synchronous Storage Datasets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems face challenges in efficiently managing and replicating data across multiple storage nodes, leading to issues with data availability, reliability, and performance.
Innovation Solution
The implementation of a storage system that utilizes asynchronous replication of synchronously replicated data, allowing for efficient data management and replication across multiple storage nodes, with features such as dual Peripheral Component Interconnect (PCI) flash storage devices and non-volatile Random Access Memory (NVRAM) for fast-write data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is used to ensure data consistency across storage nodes, then data reliability is improved, but system performance and latency are worsened
Solution Approach 1:
The patent segments the replication process into two distinct phases: synchronous replication for critical data consistency and asynchronous replication for performance optimization. This allows the system to maintain data reliability through synchronous replication while reducing latency by switching to asynchronous replication for non-critical operations, effectively dividing the replication function into specialized components.
Solution Approach 2:
The system dynamically adjusts the replication mode between synchronous and asynchronous based on operational requirements. The patent implements a flexible architecture that can switch replication strategies in real-time, allowing the system to optimize between reliability and performance depending on the specific data access patterns and system state, rather than being locked into a single static approach.
2Reliability
If data is replicated across multiple storage nodes, then data availability is improved, but system complexity is worsened
Solution Approach 1:
The patent introduces a controller as an intermediary component that manages the complexity of multi-node replication. This controller coordinates data distribution, monitors replication status, and handles failover logic, thereby centralizing the complexity management and simplifying the overall system architecture while maintaining high data availability across multiple storage nodes.
3Productivity
If fast-write data storage is implemented using NVRAM, then write performance is improved, but system cost is worsened
Solution Approach 1:
The patent applies local quality by using NVRAM only for specific fast-write operations where performance is critical, rather than uniformly across the entire storage system. This selective deployment of expensive high-performance storage media optimizes write performance for time-sensitive data while controlling overall system costs by using more economical storage solutions for less time-critical data.
Data Source
AI summary
A recovery storage system asynchronously protecting a synchronously replicated dataset, where the asynchronous protection of the synchronously replicated dataset includes: receiving, by a recovery storage system, an identifier of a synchronously replicated dataset, wherein the synchronously replicated dataset is a dataset synchronously replicated across the plurality of storage systems; asynchronously replicating, on the recovery storage system, the synchronously replicated dataset from the plurality of storage systems; detecting that each dataset in the synchronously replicated dataset on the plurality of storage systems has become unavailable; and restoring the dataset on the recovery storage system.