Fibre Channel over Ethernet Encapsulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data center networks face complexity and high costs due to increasing numbers of servers and connectivity requirements, with existing solutions like Infiniband being costly and not scalable for large enterprises, and Fibre Channel (FC) being more reliable but expensive for storage traffic.
Innovation Solution
Implementing a Low Latency Ethernet (LLE) solution that transforms FC frames into a format suitable for Ethernet transport, using multiple virtual lanes with 'drop' and 'no-drop' behaviors, and active buffer management to achieve high bandwidth and low latency at a lower cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Fibre Channel is used for storage traffic, then reliability is improved, but cost increases
Solution Approach 1:
The patent merges Fibre Channel and Ethernet networks into a unified network infrastructure. FC frames are encapsulated within Ethernet frames, allowing both storage traffic (FC) and general data traffic (Ethernet) to share the same physical network infrastructure, switches, and cables. This consolidation reduces device complexity and cost while maintaining the reliability of FC for storage traffic through protocol-level guarantees.
Solution Approach 2:
The patent introduces an encapsulation/decapsulation mechanism as an intermediary layer between FC and Ethernet protocols. This intermediary process allows FC frames to be transported over Ethernet infrastructure without requiring separate FC hardware throughout the network, reducing the need for expensive FC switches and cables while preserving FC's reliable delivery characteristics through buffer credit flow control.
2Device complexity
If Ethernet is used for storage traffic, then cost is reduced, but reliability deteriorates due to packet dropping
Solution Approach 1:
The patent changes the flow control parameters of Ethernet to match Fibre Channel's no-drop behavior. By implementing buffer credit mechanisms and adjusting drop policies, Ethernet ports handling FC traffic guarantee buffer space for incoming frames, preventing the packet dropping that normally occurs in Ethernet. This parameter change allows Ethernet to provide FC-level reliability for storage traffic while maintaining lower cost.
Solution Approach 2:
The patent introduces dynamic buffer management where buffer allocation and flow control credits are dynamically adjusted based on traffic conditions and quality of service requirements. This allows the network to adaptively guarantee bandwidth and prevent drops for storage traffic while still allowing standard Ethernet behavior for non-critical traffic, achieving reliability where needed without sacrificing cost-effectiveness.
3Reliability
If separate FC and Ethernet networks are maintained, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent merges separate FC and Ethernet networks into a unified infrastructure where FC frames are encapsulated within Ethernet frames. This allows both protocols to coexist on the same physical network, reducing the number of separate switches, cables, and network architectures needed while maintaining the reliability characteristics of FC through encapsulation and flow control mechanisms.
Solution Approach 2:
The patent makes the network infrastructure universal by enabling Ethernet switches and cables to handle both standard Ethernet traffic and encapsulated FC storage traffic. This multi-functionality eliminates the need for dedicated FC network hardware throughout the infrastructure, reducing complexity while preserving reliability through protocol-specific handling at the encapsulation layer.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A Data Center Ethernet ("DCE") network and related methods and device are provided. A DCE network simplifies the connectivity of data centers and provides a high bandwidth, low latency network for carrying Ethernet, storage and other traffic. A Fibre Channel ("FC") frame, including FC addressing information, is encapsulated in an Ethernet frame for transmission on a Data Center Ethernet ("DCE") network. The Ethernet address fields may indicate that the frame includes an FC frame, e.g., by the use of a predetermined Organization Unique Identifier ("OUI") code in the D_MAC field, but also include Ethernet addressing information. Accordingly, the encapsulated frames can be forwarded properly by switches in the DCE network whether or not these switches are configured for operation according to the FC protocol. Accordingly, only a subset of Ethernet switches in the DCE network needs to be FC-enabled. Only switches so configured will require an FC Domain_ID.