Hypervisor Volume Group Load Balancing in Clustered Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Load balancing in clustered virtualization environments is challenging due to the difficulty in configuring and managing storage controllers, especially in scale-out systems where nodes can be added or removed dynamically, leading to uneven resource utilization and performance degradation.
Innovation Solution
Implementing a method for hypervisor-attached volume group load balancing by partitioning virtual disks into shards and assigning them to storage controllers based on load distribution, using iSCSI connections to redirect requests and reassign shards as needed to maintain balanced resource usage across nodes, even as the cluster configuration changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If client-side software is used for load balancing, then resource usage can be optimized, but configuration and management of storage controllers becomes difficult in scale-out systems
Solution Approach 1:
The patent introduces a load balancer as an intermediary component that sits between clients and storage controllers. This load balancer handles the complex configuration and routing decisions, while clients simply connect to the load balancer. The load balancer maintains mapping information between clients and storage controllers, automating the configuration process and making scale-out systems manageable without requiring manual configuration of each storage controller.
2Productivity
If storage controllers are manually configured in scale-out systems, then load balancing can be implemented, but the system becomes difficult to manage when nodes are added or removed
Solution Approach 1:
The patent implements dynamic load balancing where the load balancer automatically adapts to changes in the system configuration. When nodes are added or removed, the load balancer dynamically updates its internal mappings and redistributes loads without requiring manual reconfiguration. This dynamic approach allows the system to seamlessly handle scale-out operations while maintaining load balancing efficiency.
3Ease of operation
If load balancing is implemented without automatic redistribution, then configuration is simpler, but uneven resource utilization and performance degradation occur
Solution Approach 1:
The patent implements a feedback mechanism where the load balancer continuously monitors the status and performance of storage controllers. Based on this feedback information, the load balancer automatically makes intelligent routing decisions to distribute loads evenly across available resources. This feedback-driven approach ensures optimal resource utilization while maintaining configuration simplicity, as the system self-adjusts based on real-time conditions.
Data Source
AI summary
Communication between VMs running in a clustered virtualization environment and a distributed storage system is managed. The distributed storage system consists of multiple storage/compute nodes. A hypervisor communicates with a controller in a local node on behalf of a VM. The controller will redirect a connection to a second controller on different node. The controller to which the connection is redirected is selected so as to distribute the load evenly across the cluster.