Virtual Volume Placement Based on Storage Node Activity Levels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional virtual volume placement techniques are inefficient as they do not accurately evaluate the IO workload from VM operations, leading to imbalanced resource consumption across storage processors, which affects overall system performance and storage efficiency.
Innovation Solution
The method involves determining the activity level of each storage node based on the IO access frequency probabilities of virtual volumes using an Analytic Hierarchy Process (AHP) and selecting the node with the lowest total activity level to host new virtual volumes, ensuring balanced resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional virtual volume placement techniques are used, then the placement process is simple, but the IO workload distribution becomes imbalanced across storage processors
Solution Approach 1:
The patent introduces activity level as a new parameter to evaluate storage node status, moving beyond simple volume count metrics. By calculating activity levels based on IO operations and changing placement decisions based on this dynamic parameter, the system achieves better workload distribution while maintaining operational simplicity through automated calculations.
2Productivity
If activity level-based placement is implemented, then workload distribution improves, but calculation complexity increases
Solution Approach 1:
The system performs self-service by automatically calculating activity levels and making placement decisions without external intervention. The storage system monitors its own IO operations and uses this data to dynamically adjust volume placement, eliminating the need for complex external management while improving workload distribution.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors IO operations on storage nodes and uses this information to adjust placement decisions. By feeding back activity level data to the placement algorithm, the system dynamically optimizes workload distribution based on current system state without requiring complex external control.
3Measurement precision
If activity level monitoring is performed for all storage nodes, then placement accuracy improves, but system overhead increases
Solution Approach 1:
The activity level monitoring mechanism serves multiple functions simultaneously: it tracks IO workload for placement decisions, provides system performance metrics, and enables dynamic load balancing. By making the monitoring system multi-functional, the patent reduces the need for separate overhead mechanisms while improving placement accuracy.
Data Source
AI summary
Methods, apparatus, and processor-readable storage media for placement of virtual volumes based on activity levels are provided herein. An example computer-implemented method includes obtaining a request to create a new virtual volume on a storage cluster comprising a plurality of storage nodes; determining a total activity level for each of a plurality of storage nodes of the storage cluster in response to the request, wherein the total activity level for a given storage node is determined based at least in part on an activity level value of each virtual volume currently hosted on the storage node; selecting one of the plurality of storage nodes to host the new virtual volume based at least in part on the determined total activity levels; and creating the new virtual volume on the selected storage node.