Notes of 'Designing.Storage.Area.Networks.2nd.Ed'

《Addison.Wesley.Designing.Storage.Area.Networks.2nd.Edition.chm》

 

1.1 Using the SNIA Shared Storage Model

Sharing storage resources among multiple servers or workstations requires a peer-to-peer network that joins targets to initiators. The composition of that network and the type of storage data traversing it vary from one architecture to another. Generally, shared storage architectures divide into storage area networks (SANs) and network-attached storage (NAS). For SANs, the network infrastructure may be Fibre Channel or Gigabit Ethernet, and the type of storage data being transported is block Small Computer Systems Interface (SCSI) data. For NAS, the network infrastructure is typically Ethernet (Fast Ethernet or Gigabit Ethernet), and the type of storage data carried across the network is file-based. At the most abstract level, then, the common denominator between SAN and NAS is that both enable sharing of storage resources by multiple initiators, whether block-based or file-based.

 

Chapter 2. Storage and Networking Concepts

Storage and networking have evolved as two distinct technologies. Storage, as embodied by the concepts of initiators and targets, assumes a master/slave relationship between connected devices. Networking, by contrast, assumes more egalitarian peer-to-peer relationships between connected devices. The focus of storage technology has been efficient data placement, whereas the focus of networking has been efficient data transport.

 

2.1.6 Upper-Layer Protocol Support

For Fibre Channel SANs, the upper-layer protocol is the Fibre Channel Protocol (FCP). For IP-based SANs, the upper-layer protocol may be Internet Fibre Channel Protocol (iFCP), which puts FCP over TCP/IP, or Internet SCSI (iSCSI), which wraps serial SCSI in TCP/IP. These upper-layer protocols encapsulate basic SCSI read and write commands, status, and data.

 

2.2 The SCSI Architecture

SCSI targets are identified by the operating system in a three-part bus/ target/LUN descriptor. The bus designator is one of several SCSI interfaces installed on a host system. A traditional parallel SCSI adapter card may represent a single bus, with that bus supporting multiple daisy chained disks. Alternatively, a Fibre Channel (HBA) or iSCSI network interface may be viewed by the operating system as a SCSI bus. Multiple installed cards would be seen as multiple bus numbers. The target represents a single storage resource on a bus daisy chain, whereas the LUN (logical unit number) designation identifies the SCSI client within the target. A single physical disk, for example, may have one logical unit and consequently one logical unit number. A RAID controller attached to a bus may represent a single target but have multiple logical units and multiple LUNs assigned.

 

The relationship between SCSI initiators and targets is defined in the SCSI Architectural Model (SAM-2) shown in . For networked storage, additional standards documents further define serial SCSI implementations. Serial SCSI implementations such as Fibre Channel and iSCSI are a component of the SAM-2 definitions for SCSI-3 commands.

 

Reads and writes of data between SCSI initiators and targets are performed with a series of SCSI commands, delivery requests, delivery actions, and responses. SCSI commands and parameters are specified in the command descriptor block (CDB). The CDB is part of a command frame sent from initiator to target. For improved performance on write operations, the frame may also contain data to be written to the media. Serial SCSI transport protocols such as Fibre Channel and iSCSI simply encapsulate CDBs as their payload. The CDB is encapsulated within the Fibre Channel Protocol information units (IUs), whereas in iSCSI it is carried in the iSCSI protocol data unit (PDU).

 

For direct-attached storage configurations, the limitations of parallel SCSI are being addressed by Serial Attached SCSI (SAS). The SAS specification is being developed by ANSI T10 as the next-generation direct-attached storage interface. Although SAS is not intended as a replacement for networked storage, it will provide efficiencies in local storage connectivity and will be immune to the skew and distance limitations of parallel cabling.

 

2.4 Network-Attached Storage

Although the relationship between NAS and SAN has been somewhat obscured by vendor marketing, the distinction between them is fairly straightforward. NAS delivers storage data in the form of files; SANs deliver storage data as blocks. NAS therefore uses file transport protocols such as NFS and CIFS, whereas SANs use SCSI, a block transport protocol.

 

3.1 Fibre Channel Layers

Fibre Channel is a standards-based networking architecture. The standards provide definitions for physical-layer attributes, transport controls, and upper-level support of TCP/IP, SCSI-3, (HiPPI), and other protocols. Fibre Channel is a gigabit transport, with current implementations at 1Gbps and 2Gbps. The governing body for Fibre Channel standards is the NCITS/ANSI T11X3 committee.

 

Fibre Channel standards define a multilayered architecture for moving data across the network. As listed in , these layers are numbered from FC-0 to FC-4. The top layer, FC-4, establishes the interface between Fibre Channel and upper-level applications. The serial SCSI protocol, for example, must associate Fibre Channel devices with storage resources that can be accessed by the operating system. For host bus adapters, this function is typically fulfilled by the device driver provided by the vendor. FC-3 is under development and may include facilities for data encryption and compression. The FC-2 layer defines how blocks of data handed down by the upper-level application will be segmented into sequences of frames for hand-off to the transport layers. This layer also includes various classes of services and flow control mechanisms. The lower two layers—FC-1 and FC-0—focus on the actual transport of data across the network. FC-1 provides facilities for encoding and decoding data for shipment at gigabit speeds and defines the command structure for accessing the media. FC-0 establishes standards for various media types, allowable lengths, and signaling.

 

Layer Number	Function	Content
FC-4	Upper-layer protocol interface	For example, SCSI-3, IP
FC-3	Common services	Under construction
FC-2	Data delivery	Framing, flow control, service class
FC-1	Ordered sets/byte encoding	8b/10b encoding, link controls
FC-0	Physical interface	Optical/electrical, cable plant

 

Fibre Channel's layered architecture is implemented on three transport topologies: point-to-point, arbitrated loop, and switched fabric. Point-to-point is a dedicated connection between two devices only, typically a server and a disk. Arbitrated loop is a shared medium, similar to Token Ring or Fiber Distributed Data Interface (FDDI), and employs a special superset of commands to control access to the medium by multiple devices. A Fibre Channel fabric is one or more switches providing higher-level services and switched bandwidth of 100MBps or 200MBps (200MBps and 400MBps full duplex) per port.