Data center bridging (DCB) is a set of enhancements to the Ethernet local area network communication protocol for use in data center environments, in particular for use with clustering and storage area networks.
Motivation
Ethernet is the primary network protocol in data centers for computer-to-computer communications. However, Ethernet is designed to be a best-effort network that may experience packet loss when the network or devices are busy.
In IP networks, transport reliability under the end-to-end principle is the responsibility of the transport protocols, such as the Transmission Control Protocol (TCP). One area of evolution for Ethernet is to add extensions to the existing protocol suite to provide reliability without requiring the complexity of TCP. With the move to 10 Gbit/s and faster transmission rates, there is also a desire for finer granularity in control of bandwidth allocation and to ensure it is used more effectively. These enhancements are particularly important to make Ethernet a more viable transport for storage and server cluster traffic. A primary motivation is the sensitivity of Fibre Channel over Ethernet to frame loss. The higher level goal is to use a single set of Ethernet physical devices or adapters for computers to talk to a Storage Area Network, Local Area network and InfiniBand fabric.
Approach
DCB aims, for selected traffic, to eliminate loss due to queue overflow (sometimes called lossless Ethernet) and to be able to allocate bandwidth on links. Essentially, DCB enables, to some extent, the treatment of different priorities as if they were different pipes. To meet these goals new standards are being (or have been) developed that either extend the existing set of Ethernet protocols or emulate the connectivity offered by Ethernet protocols. They are being (or have been) developed respectively by two separate standards bodies:
The Institute of Electrical and Electronics Engineers (IEEE) Data Center Bridging Task Group of the IEEE 802.1 Working Group Internet Engineering Task Force (IETF). Enabling DCB broadly on arbitrary networks with irregular topologies and without special routing may cause deadlocks, large buffering delays, unfairness and head-of-line blocking. It was suggested to use DCB to eliminate TCP slow start using approach of TCP-Bolt.
Terminology
Different terms have been used to market products based on data center bridging standards:
Data Center Ethernet (DCE) was a term trademarked by Brocade Communications Systems in 2007 but abandoned by request in 2008. DCE referred to Ethernet enhancements for the Data Center Bridging standards, and also including a Layer 2 Multipathing implementation based on the IETF's Transparent Interconnection of Lots of Links (TRILL) standard. Convergence Enhanced Ethernet or Converged Enhanced Ethernet (CEE) was defined from 2008 through January 2009 by group of including Broadcom, Brocade Communications Systems, Cisco Systems, Emulex, HP, IBM, Juniper Networks, QLogic. The ad-hoc group formed to create proposals for enhancements that enable networking protocol convergence over Ethernet, specially Fibre Channel. Proposed specifications to IEEE 802.1 working groups initially included: The Priority-based Flow Control (PFC) Version 0 Specification was submitted for use in the IEEE 802.1Qbb project, under the DCB task group of the IEEE 802.1 working group. The Enhanced Transmission Selection (ETS) Version 0 Specification was submitted for use in the IEEE 802.1Qaz project, under the DCB task group of the IEEE 802.1 working group. The Data Center Bridging eXchange (DCBX) Version 0 Specification was also submitted for use in the IEEE 802.1Qaz project.
IEEE task group
The following have been adopted as IEEE standards: