Low Latency Ethernet Switch Architecture for I/O Consolidation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data center networks face complexity and high costs due to the need for redundant connectivity and separate networks for Ethernet and Fibre Channel traffic, with existing solutions like Infiniband being costly and not scalable for large enterprises.
Innovation Solution
Implementing a Low Latency Ethernet (LLE) solution that transforms Fibre Channel frames into an Ethernet format, using multiple virtual lanes with different behaviors (drop, no-drop, and delayed drop) to manage traffic efficiently, and employing active buffer management with credit-based schemes for low latency and high reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate networks are used for Ethernet and Fibre Channel traffic, then reliability and performance for storage traffic are improved, but device complexity and cost increase
Solution Approach 1:
The patent merges Ethernet and Fibre Channel traffic into a single network infrastructure by implementing FC over Ethernet, where Fibre Channel frames are encapsulated and transmitted over Ethernet switches. This consolidation eliminates the need for separate FC and Ethernet networks, reducing device complexity and cost while maintaining storage traffic reliability through protocol-level guarantees.
Solution Approach 2:
The network device is designed to handle multiple traffic types (Ethernet and Fibre Channel) through a single unified platform. The switch can process both native Ethernet frames and encapsulated FC frames, providing multi-functionality that reduces the number of dedicated hardware components and simplifies the overall network architecture.
2Reliability
If redundant connectivity is implemented for high availability, then system reliability is improved, but cost and complexity increase
Solution Approach 1:
The patent combines redundant FC connectivity requirements into the Ethernet network by implementing FC over Ethernet with redundant Ethernet links. This allows high availability to be achieved using existing Ethernet redundancy mechanisms while eliminating the need for separate redundant FC infrastructure, thereby reducing overall complexity and cost.
3Reliability
If Fibre Channel is used for storage traffic, then reliability is improved, but cost and scalability decrease
Solution Approach 1:
The patent replaces expensive Fibre Channel hardware with more cost-effective Ethernet components. By encapsulating FC frames in Ethernet packets, the system can use standard Ethernet switches, cables, and interfaces instead of specialized FC hardware, significantly reducing network cost while maintaining the reliability characteristics needed for storage traffic through protocol-level flow control and error handling.
4Device complexity
If Ethernet is used for storage traffic, then cost is reduced, but latency and packet loss increase
Solution Approach 1:
The patent modifies Ethernet protocol parameters and introduces virtual lane differentiation to optimize performance for storage traffic. By creating dedicated virtual lanes with prioritized processing and adjusted buffer management, the system reduces latency and packet loss for FC-over-Ethernet traffic compared to standard Ethernet usage, while maintaining the cost advantages of Ethernet infrastructure.
Data Source
AI summary
The present invention provides methods and devices for implementing a Low Latency Ethernet (“LLE”) solution, also referred to herein as a Data Center Ethernet (“DCE”) solution, which simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet and storage traffic. Some aspects of the invention involve transforming FC frames into a format suitable for transport on an Ethernet. Some preferred implementations of the invention implement multiple virtual lanes (“VLs”) in a single physical connection of a data center or similar network. Some VLs are “drop” VLs, with Ethernet-like behavior, and others are “no-drop” lanes with FC-like behavior. Some preferred implementations of the invention provide guaranteed bandwidth based on credits and VL. Active buffer management allows for both high reliability and low latency while using small frame buffers. Preferably, the rules for active buffer management are different for drop and no drop VLs.