Deployment Guide

Data center bridging satises the needs of the following types of data center trac in a unied fabric:

• LAN trac consists of a large number of ows that are generally insensitive to latency requirements, while certain applications, such as

streaming video, are more sensitive to latency. Ethernet functions as a best-eort network that may drop packets in case of network

congestion. IP networks rely on transport protocols (for example, TCP) for reliable data transmission with the associated cost of

greater processing overhead and performance impact.

• Storage trac based on Fibre Channel media uses the SCSI protocol for data transfer. This trac typically consists of large data

packets with a payload of 2K bytes that cannot recover from frame loss. To successfully transport storage trac, data center Ethernet

must provide no-drop service with lossless links.

• Servers use InterProcess Communication (IPC) trac within high-performance computing clusters to share information. Server trac

is extremely sensitive to latency requirements.

To ensure lossless delivery and latency-sensitive scheduling of storage and service trac and I/O convergence of LAN, storage, and server

trac over a unied fabric, IEEE data center bridging adds the following extensions to a classical Ethernet network:

• 802.1Qbb - Priority-based Flow Control (PFC)

• 802.1Qaz - Enhanced Transmission Selection (ETS)

• 802.1Qau - Congestion Notication

• Data Center Bridging Exchange (DCBx) protocol

NOTE: In Dell Networking OS version 9.4.0.x, only the PFC, ETS, and DCBx features are supported in data center bridging.

Priority-Based Flow Control

In a data center network, priority-based ow control (PFC) manages large bursts of one trac type in multiprotocol links so that it does not

aect other trac types and no frames are lost due to congestion.

When PFC detects congestion on a queue for a specied priority, it sends a pause frame for the 802.1p priority trac to the transmitting

device. In this way, PFC ensures that large amounts of queued LAN trac do not cause storage trac to be dropped, and that storage

trac does not result in high latency for high-performance computing (HPC) trac between servers.

PFC enhances the existing 802.3x pause and 802.1p priority capabilities to enable ow control based on 802.1p priorities (classes of

service). Instead of stopping all trac on a link (as performed by the traditional Ethernet pause mechanism), PFC pauses trac on a link

according to the 802.1p priority set on a trac type. You can create lossless ows for storage and server trac while allowing for loss in

case of LAN trac congestion on the same physical interface.

The following illustration shows how PFC handles trac congestion by pausing the transmission of incoming trac with dot1p priority 3.

Figure 1. Priority-Based Flow Control

In the system, PFC is implemented as follows:

• PFC is supported on specied 802.1p priority trac (dot1p 0 to 7) and is congured per interface. However, only two lossless queues

are supported on an interface: one for Fibre Channel over Ethernet (FCoE) converged trac and one for Internet Small Computer

System Interface (iSCSI) storage trac. Congure the same lossless queues on all ports.

Data Center Bridging (DCB)