Front-Back End Virtualization Engine Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage area networks face limitations in efficiently and effectively configuring virtualization mechanisms, leading to resource inefficiencies and bottlenecks, particularly in fibre channel fabrics, where virtualization engines can become overwhelmed during high transaction volumes, disrupting network performance.
Innovation Solution
The implementation of a virtualization engine architecture that separates tasks into front end and back end virtualization engines, where the front end performs light processing and selects a back end engine for resource-intensive tasks like virtual to physical address mapping, allowing dynamic scaling and load balancing without requiring host or target reconfiguration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtualization engines are used to perform address mapping and redundancy operations in storage area networks, then resource efficiency and data protection are improved, but the virtualization engines become overwhelmed during high transaction volumes, causing network performance degradation
Solution Approach 1:
The virtualization engine is segmented into multiple specialized components: front-end virtualization engines that handle I/O request reception and initial processing, back-end virtualization engines that perform resource-intensive address mapping, and load balancers that distribute workloads. This segmentation prevents any single engine from becoming overwhelmed and maintains network performance during high transaction volumes.
Solution Approach 2:
The system implements dynamic load balancing that continuously monitors virtualization engine workload and dynamically redirects I/O requests to optimize performance. The load balancer can dynamically adjust the distribution of requests across multiple back-end engines based on current resource availability and transaction volumes, preventing bottlenecks and maintaining reliability under varying loads.
2Adaptability or versatility
If virtualization resources are dynamically added to handle increased transaction volumes, then system capacity and flexibility are improved, but configuration complexity and deployment difficulty increase
Solution Approach 1:
The load balancer serves multiple functions: it acts as a configuration manager that automatically provisions new virtualization engines, a traffic distributor that balances workloads, and a monitoring system that tracks engine health and performance. This multi-functionality reduces configuration complexity by consolidating management tasks into a single universal component that handles both dynamic resource addition and ongoing optimization.
Solution Approach 2:
The system implements self-service capabilities where the load balancer automatically detects when new virtualization engines are added to the network and automatically configures them to receive I/O requests. This eliminates manual configuration requirements and allows the system to dynamically adapt to changing resource availability without increasing operational complexity.
3Speed
If address mapping operations are performed at the network level in fibre channel fabrics, then data access performance and storage resource allocation are improved, but the virtualization mechanisms become difficult to configure and maintain
Solution Approach 1:
The load balancer acts as an intermediary between the front-end virtualization engines and back-end address mapping engines. It simplifies the configuration process by abstracting the complex address mapping operations behind a simple request routing interface, allowing fast data access to be maintained while reducing the operational complexity of configuring and maintaining the virtualization mechanisms.
Data Source
AI summary
Methods and apparatus are provided for improving network virtualization in a storage area network. A virtualization engine is divided into a front end virtualization engine and a back end virtualization engine cluster. The front end virtualization is associated with one or more virtual enclosure ports. The back end virtualization engine cluster performs resource intensive virtual address to physical address mapping. Virtualization engine resources can be dynamically added to a back end virtualization engine cluster while presenting a consistent interface to hosts connected to a storage area network.