Controller-Based Volume Migration Traffic Duplication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network-based traffic duplication processes are unreliable and cannot be used for reliable data copying during volume migration or replication, which is critical for maintaining data integrity across storage systems.
Innovation Solution
A method and system that utilize port mirroring to monitor and control the duplication of traffic between host computers and storage systems, ensuring that only successful updates are reflected in the destination volume during migration or replication, and reconstructing missed updates by detecting gaps in the duplicate traffic metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network-based traffic duplication processes are used for copying volume content, then the migration process can proceed, but the reliability of data copying deteriorates
Solution Approach 1:
A dedicated controller is introduced as an intermediary between the storage systems and the network switch. This controller receives duplicate traffic from the switch, validates its integrity, and selectively forwards only successful updates to the destination volume. The controller acts as a mediator that filters out corrupted or lost packets, ensuring reliable data copying while maintaining migration productivity.
Solution Approach 2:
The system implements feedback mechanisms where the controller monitors duplicate traffic and validates updates before applying them to the destination volume. The controller tracks which updates were successfully received and applied, and can request retransmission of missed updates. This feedback loop ensures that only reliable data is copied, resolving the contradiction between migration speed and data integrity.
2Reliability
If port mirroring is used to monitor traffic, then data integrity can be maintained, but device complexity increases
Solution Approach 1:
The controller performs multiple functions: it receives duplicate traffic from the switch, validates traffic integrity, filters successful updates, and forwards them to the destination volume. By consolidating these multiple functions into a single controller device, the system maintains data integrity through port mirroring while minimizing the increase in overall system complexity.
Solution Approach 2:
The system uses port mirroring to create a copy of the network traffic flow. Instead of modifying the original traffic paths or requiring complex coordination between multiple devices, the controller simply copies the traffic information from the switch's mirror port and processes it independently. This copying approach maintains integrity without significantly increasing complexity.
3Stability of the object's composition
If all write requests are duplicated during migration, then data consistency is improved, but processing time increases
Solution Approach 1:
Instead of processing every single write request duplicately, the controller uses port mirroring to receive only the essential duplicate traffic information from the switch. This partial action approach allows the system to maintain data consistency by monitoring traffic patterns while reducing processing time by avoiding redundant validation of every individual write operation.
Solution Approach 2:
The system replaces complex mechanical traffic validation mechanisms with network-based port mirroring. Instead of requiring the controller to actively validate every packet through complex processing, the switch performs traffic duplication at the network layer, and the controller simply receives and processes this pre-filtered traffic information, significantly reducing processing time while maintaining consistency.
Data Source
AI summary
A method for updating a destination volume, the method may include monitoring duplicate traffic that represents original traffic that is exchanged between at least one host computer and a first storage system; wherein the monitoring is executed by a controller during an execution of a migration or a replication of a source volume of the first storage system to a destination volume of a second storage system; wherein the controller is not included in the first storage system; wherein the original traffic is related to the source volume; wherein the duplicate traffic is sent toward the controller by at least one switch that attempts to duplicate the original traffic; and controlling, based upon the monitoring, a process of updating the destination volume with a set of source volume updates, wherein the set of source volume updates are reflected in the duplicate traffic and are received during the execution of the migration.


