Temporary Volume Bandwidth Estimation for Storage Replication QoS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating bandwidth and providing Quality of Service (QoS) parameters in storage area networks are inefficient, as they often consume available bandwidth, affecting performance and do not accurately estimate actual traffic conditions, especially in scenarios with multiple replication partners.
Innovation Solution
A method that creates a temporary volume and sends a sequential workload of actual traffic between storage devices to estimate QoS parameters, allowing users to modify and control data flow based on these parameters, thereby preventing bandwidth saturation and enabling differentiated QoS settings between replication partners.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If bandwidth estimation methods consume available bandwidth during replication, then replication can proceed, but performance of other applications is adversely affected
Solution Approach 1:
The system performs bandwidth estimation and QoS parameter measurement before actual replication operations begin. By creating temporary volumes and sending sequential workloads in advance, the system establishes baseline performance metrics without interfering with production replication traffic, thus preventing performance degradation of other applications while enabling informed replication planning
Solution Approach 2:
The system introduces temporary volumes as intermediary structures that facilitate bandwidth estimation without directly competing with production replication traffic. These temporary volumes serve as testbeds for measuring QoS parameters, allowing the system to gather performance data while isolating measurement activities from production workloads, thereby protecting application performance
2Ease of operation
If QoS parameters are not accurately estimated, then replication can proceed without detailed planning, but bandwidth saturation occurs affecting other applications
Solution Approach 1:
The system implements feedback mechanisms that continuously monitor actual traffic conditions and QoS parameter performance. By measuring bandwidth utilization, latency, and other metrics during and after replication operations, the system adjusts replication parameters dynamically to prevent bandwidth saturation while maintaining simple operation for users through automated control
Solution Approach 2:
The system performs preliminary QoS parameter estimation by sending sequential workloads through temporary volumes before production replication begins. This advance measurement provides accurate bandwidth and performance data that guides replication configuration, ensuring reliable bandwidth management while keeping the setup process simple through automated parameter determination
3Reliability
If dedicated links are used for replication, then replication performance is guaranteed, but bandwidth availability for other applications is reduced
Solution Approach 1:
The system implements dynamic QoS parameter adjustment that adapts replication bandwidth allocation based on actual traffic conditions and application needs. By continuously monitoring network utilization and adjusting replication rates in real-time, the system guarantees minimum replication performance while maximizing bandwidth utilization efficiency, eliminating the need for static dedicated links
Data Source
AI summary
The present invention is directed to a method and information handling system (IHS) for estimating bandwidth and providing one or more Quality of Service (QoS) parameters between a first set of one or more storage devices and at least one other set of one or more storage devices. The method comprises the following, for a given set of the at least one other set of one or more storage devices. The method creates a given temporary volume associated with the given set. The method sends a given sequential workload of actual traffic between the first set and the given set. The method estimates, based upon the results of sending the given sequential workload of actual traffic, the one or more Quality of Service (QoS) parameters for the given set.