Service Manual

11
Data Center Bridging (DCB)
Ethernet Enhancements in Data Center Bridging
The following section describes DCB.
The device supports the following DCB features:
• Priority-based ow control (PFC)
• Enhanced transmission selection (ETS)
DCB refers to a set of IEEE Ethernet enhancements that provide data centers with a single, robust, converged network to support
multiple trac types, including local area network (LAN), server, and storage trac. Through network consolidation, DCB results in
reduced operational cost, simplied management, and easy scalability by avoiding the need to deploy separate application-specic
networks.
For example, instead of deploying an Ethernet network for LAN trac, include additional storage area networks (SANs) to ensure
lossless Fibre Channel trac, and a separate InniBand network for high-performance inter-processor computing within server
clusters, only one DCB-enabled network is required in a data center. The Dell Networking switches that support a unied fabric and
consolidate multiple network infrastructures use a single input/output (I/O) device called a converged network adapter (CNA).
A CNA is a computer input/output device that combines the functionality of a host bus adapter (HBA) with a network interface
controller (NIC). Multiple adapters on dierent devices for several trac types are no longer required.
Data center bridging satises the needs of the following types of data center trac in a unied fabric:
LAN
trac LAN trac consists of many ows that are 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 the 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 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 Storage trac based on Fibre Channel media uses the Small Computer System Interface (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.
InterProcess
Communication
(IPC) trac
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)
Data Center Bridging (DCB)
207