Storage Volume Set Performance Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional file systems lack the ability to configure intermediate levels of performance, such as groups of overlapping volume sets, making it difficult to manage workloads that exceed initial configurations, and may lead to performance violations when trying to adjust to higher demands.
Innovation Solution
A system and method that allow administrators to configure subsets of storage units with specific input/output and bit rate limits, enabling dynamic adjustments while adhering to a performance policy, allowing for configuration at various levels like volumes, volume sets, domains, and host sets, ensuring stable performance across the system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If performance parameters are specified only at the highest level (file system level) or lowest level (single volume level), then configuration simplicity is maintained, but the ability to manage intermediate workload groups and adapt to specific performance needs is limited
Solution Approach 1:
The patent segments the storage system into multiple hierarchical levels (file system level, volume set level, and individual volume level), allowing performance parameters to be configured at each level independently. This enables granular control over intermediate workload groups while maintaining overall system coherence, resolving the contradiction between adaptability and complexity.
Solution Approach 2:
The patent introduces a new dimensional layer (volume set level) between the traditional highest level (file system) and lowest level (individual volume), creating a multi-dimensional configuration space. This additional dimension allows administrators to configure performance parameters for specific groups of volumes without affecting the entire system or requiring individual volume configuration, thus enhancing adaptability without proportionally increasing complexity.
2Adaptability or versatility
If the file system is reconfigured to cope with higher future workloads, then future performance demands may be met, but it becomes hard to predict future workload demands and configure appropriately
Solution Approach 1:
The patent implements dynamic performance configuration at multiple levels, allowing administrators to adjust performance parameters for specific volume sets based on actual workload patterns rather than making static predictions for the entire system. This enables adaptive response to changing workload demands without requiring upfront prediction of future requirements.
Solution Approach 2:
The patent enables different performance configurations for different volume sets based on their specific workload characteristics. Instead of uniformly reconfiguring the entire file system to handle potential future workloads, administrators can apply localized performance adjustments to specific volume sets experiencing higher demands, reducing the time and uncertainty associated with predicting and preparing for future workload changes.
3Productivity
If individual storage units are configured with higher I/O and bit rate limits, then they can handle higher workloads, but the rest of the system may violate performance policies
Solution Approach 1:
The patent segments the storage system into manageable volume sets with configurable performance parameters at each level. This allows high-performance configurations for specific volume sets handling heavy workloads while maintaining policy-compliant configurations for other volume sets, thus achieving both high productivity for critical workloads and overall system reliability through differentiated configuration management.
Solution Approach 2:
The patent applies the principle of local quality by allowing different performance limits to be configured for different volume sets based on their specific requirements. High-I/O volume sets can be configured with higher throughput limits to handle heavy workloads, while other volume sets maintain lower, policy-compliant limits, ensuring that local performance optimizations do not cause global performance policy violations.
Data Source
AI summary
Disclosed herein are techniques for managing the performance of a storage system. A subset of a plurality of storage units is associated with a rule that specifies a number of input and output transactions and a number of bits per time unit. The associations are adjusted in accordance with a performance policy and changes in the subset of the plurality of storage units.