N-way router synchronous replication heterogeneous storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In heterogeneous storage networks with mixed node performance, synchronous replication often degrades to the level of slower storage nodes, leading to increased costs or compromised RPO and RTO metrics, as high-cost high-performance nodes are required for synchronous replication or asynchronous replication is used to maintain performance.
Innovation Solution
Implementing a programmable n-way router to route I/Os efficiently, initially storing data sequentially on low-cost high-capacity hard disks for rapid completion and later scattering it to proper locations, and packetizing I/Os to reduce interrupts and context switches, allowing for synchronous replication with low RPO and RTO while leveraging low-cost backup storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If synchronous replication is implemented between heterogeneous storage nodes (SAS and SATA), then data consistency and low RPO/RTO are achieved, but system performance degrades to the level of slower SATA nodes
Solution Approach 1:
The patent segments the replication path by introducing an intermediary buffer node that separates the SAS primary node from the SATA secondary node. The buffer node receives data from the SAS node, stores it temporarily, and then transfers it to the SATA node. This segmentation allows the fast SAS node to maintain high performance while the SATA node operates at its own speed, preventing the slow node from bottlenecking the entire system.
Solution Approach 2:
The patent introduces a buffer node as an intermediary between the primary SAS storage node and the secondary SATA storage node. This intermediary buffer absorbs the performance difference between the two heterogeneous nodes, allowing synchronous replication to proceed without the slow SATA node degrading the overall system performance. The buffer node acts as a mediator that decouples the performance requirements of the two nodes.
2Reliability
If high-performance SAS nodes are used for synchronous replication, then low RPO and RTO metrics are achieved, but system cost increases
Solution Approach 1:
The patent applies local quality by using different storage node types for different functions: the SAS node handles high-performance primary storage operations, while the SATA node handles lower-cost backup storage. The buffer node locally manages the data transfer and timing, allowing each node to operate in its optimal performance range while maintaining synchronous replication guarantees.
Solution Approach 2:
The patent changes the operational parameters of the replication system by introducing a buffer that allows asynchronous data transfer between nodes while maintaining synchronous replication semantics. The buffer node adjusts transfer timing and pacing, changing the effective parameters of data flow to accommodate heterogeneous node performance while maintaining low RPO/RTO metrics.
3Productivity
If asynchronous replication is used between heterogeneous nodes, then system performance is maintained, but RPO and RTO metrics deteriorate
Solution Approach 1:
The patent introduces dynamic buffering and timing control at the buffer node to adapt to varying data flow rates between SAS and SATA nodes. The buffer dynamically adjusts its operations to ensure data is available synchronously when needed, maintaining low RPO/RTO metrics while allowing the underlying physical transfer to accommodate performance differences.
Data Source
AI summary
Technologies for efficient synchronous replication across heterogeneous storage nodes can provide the performance of high-speed storage units while leveraging low-cost and high-capacity backup storage units within the same system. The performance of low-cost, high-capacity hard disks may be improved by initially storing data into sequential physical locations. A sequential journal of I/Os may be used in a replicated secondary node to allow for rapid completion of I/Os. A separate background process can later scatter the sequentially logged I/O data into its proper location for storage. A programmable n-way router can be configured to route I/Os as needed to improve overall performance of the storage unit. A secondary node log device can also be used to provide continuous data protection (CDP). Lastly, packetizing together I/Os prior to delivery to a secondary node may reduce interrupts and context switches in the primary node, thereby improving performance of the storage system.