Logical Storage Access Balancing with Dynamic ALUA and ANA States
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage systems face challenges in dynamically adjusting asymmetric access state settings due to varying path performance characteristics and temporary load imbalances between storage controllers, particularly in metro/stretched configurations where host devices are restricted in adjusting preferences between local and remote storage systems.
Innovation Solution
Implementing a processing device to identify logical storage devices across multiple storage systems, measure response times, and dynamically modify asymmetric access state settings such as ALUA or ANA states based on these measurements to optimize path selection and load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If asymmetric access state settings are statically configured to prefer local storage system paths, then access speed to local storage is improved, but load balancing between storage controllers deteriorates and remote storage paths cannot be utilized when local storage is overloaded
Solution Approach 1:
The patent implements dynamic modification of asymmetric access state settings (ALUA/ANA) based on real-time response time measurements. The host device continuously monitors storage system performance and adjusts path preferences dynamically, transitioning from static configuration to adaptive behavior that responds to changing load conditions across storage controllers.
Solution Approach 2:
The system employs feedback mechanisms by measuring response times for IO operations to both local and remote storage systems, then using this measurement data to inform subsequent asymmetric access state configuration decisions. This closed-loop control enables the system to adapt to varying storage controller loads automatically.
2Stability of the object's composition
If host device is restricted from adjusting asymmetric access state settings, then storage system stability is improved, but performance optimization under varying load conditions deteriorates
Solution Approach 1:
The host device autonomously performs asymmetric access state setting modifications based on its own measurements of storage system performance. The host self-adjusts path preferences without requiring external intervention or complex coordination with storage controllers, enabling rapid adaptation while maintaining system stability through decentralized control.
Solution Approach 2:
The system performs preliminary measurements of response times and evaluates storage controller performance before making asymmetric access state configuration changes. This preparatory assessment ensures that modifications are based on accurate performance data, reducing the risk of suboptimal configurations.
3Productivity
If paths to remote storage system are used, then load balancing is improved, but access latency increases due to longer path lengths
Solution Approach 1:
The system dynamically changes the asymmetric access state parameters (AO/ANO designations) based on measured response times. When remote storage controllers exhibit better performance characteristics, the system adjusts parameters to favor those paths, effectively treating the faster remote path as the optimized access route despite its longer physical distance.
Solution Approach 2:
The patent leverages asymmetric access state settings to create different access preferences for different paths. By dynamically adjusting which paths are marked as optimized (AO) versus non-optimized (ANO), the system can asymmetrically route IO operations through the most efficient paths regardless of their physical characteristics, balancing load while minimizing latency.
Data Source
AI summary
An apparatus in one embodiment includes at least one processing device comprising a processor coupled to a memory. The at least one processing device is configured to identify a logical storage device accessible in at least first and second storage systems, to measure response times for accessing the logical storage device in each of the first and second storage systems, and to modify, based at least in part on the measured response times, asymmetric access state settings for the logical storage device. Modifying asymmetric access state settings for the logical storage device in some embodiments illustratively comprises modifying asymmetric logical unit access (ALUA) or asymmetric namespace access (ANA) state settings for the logical storage device. Such modifications are illustratively performed in a dynamic manner that responds to variations in the measured response times so as to facilitate load balancing across storage controllers of the first and second storage systems.