Storage Address QoS Control for LUN Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing QoS control methods in storage systems waste resources and provide lower quality of service by allocating high levels of system resources to entire LUNs, even though only specific addresses are accessed, leading to reduced performance in other LUNs.
Innovation Solution
The method involves acquiring the storage address corresponding to a data object within a storage device and performing QoS control only on that address, rather than the entire LUN, allowing for more precise resource allocation and improved service quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If QoS control is performed on the whole LUN, then the access quality to the data object is improved, but the system resources are wasted
Solution Approach 1:
The patent segments the LUN into smaller address units, enabling QoS control to be applied to specific storage addresses rather than the entire LUN. This segmentation allows precise resource allocation to only the addresses that need QoS control, avoiding waste on other addresses in the same LUN.
Solution Approach 2:
The patent implements local quality control by applying QoS parameters to specific storage addresses where data objects are located, rather than uniformly across the entire LUN. This allows different QoS levels to be applied locally to different addresses based on actual access needs.
2Reliability
If system resources are allocated to a certain LUN, then the access quality to data in that LUN is improved, but resources allocated to other LUNs are reduced
Solution Approach 1:
By segmenting the QoS control unit from LUN-level to address-level, the patent enables finer-grained resource allocation. This allows resources to be dynamically allocated to specific addresses across different LUNs based on actual needs, improving overall system productivity while maintaining access quality where required.
Solution Approach 2:
The patent changes the control parameter from LUN-level QoS settings to address-level QoS settings. This parameter change enables more flexible and efficient resource distribution across multiple LUNs, as resources can be allocated to specific addresses rather than being locked at the LUN level.
3Reliability
If QoS control is performed on the whole LUN, then the user requirements are satisfied, but the resource allocation precision is reduced
Solution Approach 1:
The patent segments the control granularity from LUN-level to individual storage address-level, dramatically improving resource allocation precision. This allows resources to be allocated exactly where needed in the address space, rather than broadly across entire LUNs.
Solution Approach 2:
The patent applies local quality control by setting QoS parameters for specific storage addresses where data objects reside. This ensures user requirements are satisfied with precise resource allocation to the exact locations needed, rather than allocating resources across entire LUNs.
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
The present invention provides a method, an apparatus, and a system for controlling quality of service of a storage system. The method includes: acquiring a preset data object, and acquiring a storage address, in a storage device, corresponding to the data object according to the preset data object, where the data object includes an application program or a file, and the acquiring the storage address, in the storage device, corresponding to the data object includes: acquiring the storage address, in the storage device, corresponding to the data object stored in a file system by searching the file system; and sending the storage address to the storage device, so that the storage device performs quality of service control on access to the storage address. The present invention solves the problems of resource waste and lower quality of service in a QoS control manner, saves system resources of the storage system, improves the quality of service of the storage system, and can satisfy user requirements more accurately.