NPIV Proxy Gateway Engine for Virtual Fabric Login
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information handling systems require multiple converged network adapters or host bus adapters for each virtual machine to log into separate virtual fabrics, leading to under-utilization of bandwidth and excess hardware, as they can only operate via one fabric at a time.
Innovation Solution
An Information Handling System (IHS) with a processing system and memory that includes an N_Port ID Virtualization (NPIV) Proxy Gateway (NPG) engine, which receives traffic from virtual machines, determines the associated virtual fabric identifier, tags the traffic, and transmits it to the Fibre Channel networking device, allowing multiple virtual machines to log into different virtual fabrics through a single physical adapter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple converged network adapters or host bus adapters are used for each virtual machine to log into separate virtual fabrics, then each virtual machine can access its designated virtual fabric, but the system requires excess hardware and under-utilizes bandwidth on the adapters
Solution Approach 1:
The patent merges multiple virtual fabric access capabilities into a single physical adapter by implementing virtualization at the adapter level. The host bus adapter is virtualized to present multiple virtual adapters to virtual machines, allowing each VM to have dedicated access to different virtual fabrics while sharing the physical hardware resource. This eliminates the need for multiple physical adapters while maintaining fabric isolation and security.
Solution Approach 2:
The physical host bus adapter is designed to perform multiple functions by supporting simultaneous connections to multiple virtual fabrics through virtualization. A single adapter can handle traffic for multiple virtual machines on different virtual fabrics, making the hardware universal and eliminating under-utilization. The adapter dynamically routes traffic based on virtual machine identifiers and fabric mappings.
2Device complexity
If a single physical adapter is used for multiple virtual machines to access different virtual fabrics, then hardware requirements are reduced and bandwidth utilization is improved, but the system must manage complex adapter virtualization and fabric mapping
Solution Approach 1:
The virtualization layer automatically performs fabric mapping and adapter configuration without requiring manual intervention. The system self-manages the mapping between virtual machines, virtual adapters, and virtual fabrics by monitoring traffic patterns and dynamically assigning fabric connections. This eliminates the operational complexity of manual configuration while maintaining secure fabric isolation.
Solution Approach 2:
The patent introduces a virtualization intermediary layer between the physical adapter and virtual machines that simplifies management. This intermediary handles fabric mapping, traffic routing, and authentication automatically. It translates virtual machine requests into appropriate physical adapter operations, shielding administrators from the underlying complexity while enabling efficient resource utilization.
3Ease of operation
If multiple adapters are used for each virtual fabric, then fabric access is simplified for each virtual machine, but bandwidth on individual adapters is under-utilized due to dedicated assignment
Solution Approach 1:
The patent implements dynamic fabric assignment where virtual machine connections to virtual fabrics are not static but can be adjusted based on workload requirements. The virtualized adapter layer dynamically routes traffic between different virtual fabrics as needed, allowing bandwidth to be reallocated in real-time. This maintains simple fabric access for each VM while optimizing overall bandwidth utilization across the physical adapter.
Solution Approach 2:
The physical adapter maintains continuous useful action by keeping the physical connection active and utilizing it for multiple virtual fabric traffic streams simultaneously. Instead of dedicating the adapter to a single fabric (which would leave it idle for other fabrics), the system continuously processes traffic from multiple virtual machines on different fabrics through the same physical interface, maximizing bandwidth utilization without compromising access simplicity.
Data Source
AI summary
A virtual fabric login system includes a FC networking device that provides access to a plurality of virtual fabrics, a server device that provides a plurality of virtual machines, and a FCF device that is coupled to a physical server device port included on the server device and coupled to the FC networking device. The FCF device receives, via the physical server device port, first traffic from a first virtual machine and includes a first virtual machine identifier that identifies the first virtual machine. The FCF device determines, using a virtual-machine-to-virtual-fabric mapping, that the first virtual machine identifier is associated with a first virtual fabric identifier that identifies a first virtual fabric included in the plurality of virtual fabrics and tags the first traffic with the first virtual fabric identifier. The FCF device then transmits the tagged first traffic to the FC networking device.