Virtual Splitter Latency-Based Replication Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data protection systems face challenges in maintaining continuous data protection by keeping pace with high data transaction rates at production sites without slowing them down, leading to potential backlogs and system shutdowns due to the overhead of journaling processes.
Innovation Solution
A system that monitors latency between splitters and replication appliances in a replication cluster, dynamically assigning consistency groups to the appliance with the lowest latency to minimize IO latency and prevent backlog buildup.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data replication uses journaling to keep pace with production site transactions, then data protection reliability is improved, but system latency increases and productivity deteriorates
Solution Approach 1:
The system dynamically selects replication appliances based on real-time latency measurements. The splitter monitors multiple replication appliances and adapts its routing decisions according to current performance conditions, allowing the system to respond to changing load conditions and maintain optimal performance
Solution Approach 2:
The splitter acts as an intermediary component that sits between the production storage system and multiple replication appliances. It intercepts journaling traffic and intelligently routes it to the most appropriate replication appliance, preventing the need for the production system to directly manage multiple replication targets and reducing overall system latency
2Reliability
If data replication processes all transactions at the backup site, then data protection completeness is improved, but backup site productivity deteriorates due to backlog
Solution Approach 1:
The system segments the replication workload by distributing journaling traffic across multiple replication appliances instead of concentrating all transactions on a single backup site. This segmentation allows parallel processing of replication tasks and prevents any single appliance from becoming a bottleneck
Solution Approach 2:
The splitter dynamically distributes replication traffic based on the current capacity and latency characteristics of each replication appliance. When one appliance becomes overwhelmed, the system automatically routes additional traffic to other available appliances, maintaining overall replication throughput and preventing backlog accumulation
3Reliability
If the production site is forced to slow down to match backup capacity, then replication reliability is improved, but production site productivity deteriorates
Solution Approach 1:
The system dynamically adjusts replication routing without requiring the production site to change its transaction rate. By monitoring latency and automatically selecting the most responsive replication appliance, the system absorbs variability in backup capacity without impacting production performance
Solution Approach 2:
Multiple replication appliances provide redundant capacity that can handle varying loads. The system can route transactions to any available appliance based on current conditions, creating a universal replication infrastructure that adapts to changing demands without constraining the production site
Data Source
AI summary
A system, program product, and computer implemented method for replicating a consistency group comprising monitoring the latency between one or more splitters of the consistency group and replication appliances in the replication cluster for the consistency group (CG); wherein each replication appliance of the replication appliances is configured to be able to receive IO from the one or more splitters, determining which replication appliance of the replication appliances has the lowest latency after including additional latency resulting from assignment of the CG to the replication appliance, and configuring the splitter to replicate IO from the CG to the replication appliance determined to have the lowest latency.