iSCSI Components: An initiator (host), target (storage or iSCSI gateway), and an IP-based network are the key iSCSI components. If an iSCSI-capable storage array is deployed, then a host with the iSCSI initiator can directly communicate with the storage array over an IP network. However, in an implementation that uses an existing FC array for iSCSI communication, an iSCSI gateway is used. These devices perform the translation of IP packets to FC frames and vice versa, thereby bridging the connectivity between the IP and FC environments.
A standard NIC with software iSCSI initiator, a TCP offload engine (TOE) NIC with software iSCSI initiator, and an iSCSI HBAare the three iSCSI host connectivity options. The function of the iSCSI initiator is to route the SCSI commands over an IP network.
A standard NIC with a software iSCSI initiator is the simplest and least expensive connectivity option. It is easy to implement because most servers come with at least one, and in many cases two, embedded NICs. It requires only a software initiator foriSCSI functionality. Because NICs provide standard IP function, encapsulation of SCSI into IP packets and decapsulation are carried out by the host CPU. This places additional overhead on the host CPU. If a standard NIC is used in heavy I/O load situations, the host CPU might become a bottleneck. TOE NIC helps alleviate this burden. A TOE NIC offloads TCP management functions from the host and leaves only the iSCSI functionality to the host processor. The host passes the iSCSI information to the TOE card, and the TOE card sends the information to the destination using TCP/IP. Although this solution improves performance, the iSCSI functionality is still handled by a software initiator that requires host CPU cycles.
An iSCSI HBA is capable of providing performance benefits because it offloads the entire iSCSI and TCP/IP processing from the host processor. The use of an iSCSI HBA is also the simplest way to boot hosts from a SAN environment via iSCSI. If there is no iSCSI HBA, modifications must be made to the basic operating system to boot a host from the storage devices because the NIC needs to obtain an IP address before the operating system loads. The functionality of an iSCSI HBA is similar to the functionality of an FC HBA.
iSCSI Names:
- iSCSI Qualified Name (IQN): An organization must own a registered domain name to generate iSCSI Qualified Names. This domain name does not need to be active or resolve to an address. It just needs to be reserved to prevent other organizations from using the same domain name to generate iSCSI names. A date is included in the name to avoid potential conflicts caused by the transfer of domain names. An example of an IQN is iqn.2008-02.com.example:optional_string.
The optional_string
provides a serial number, an asset number, or any other device
identifiers. An iSCSI Qualified Name enables storage administrators to
assign meaningful names to iSCSI devices, and therefore, manage those
devices more easily.
- Extended Unique Identifier (EUI): An EUI is a globally unique identifier based on the IEEE EUI-64 naming standard. An EUI is composed of the eui prefix followed by a 16-character hexadecimal name, such as eui.0300732A32598D26.
The FCIP standard has
rapidly gained acceptance as a manageable, cost-effective way to blend
the best of the two worlds: FC SAN and the proven, widely deployed IP
infrastructure. As a result, organizations now have a better way to
store, protect and move their data by leveraging investments in their
existing IP infrastructure. FCIP is extensively used in disaster
recovery implementations in which data is duplicated to the storage
located at a remote site.
FCoE
Data centers
typically have multiple networks to handle various types of I/O
traffic—for example, an Ethernet network for TCP/IP communication and an
FC network for FC communication. TCP/IP is typically used for
client-server communication, data backup, infrastructure management
communication, and so on. FC is typically used for moving block-level
data between storage and servers. To support multiple networks, servers
in a data center are equipped with multiple redundant physical network
interfaces—for example, multiple Ethernet and FC cards/adapters. In
addition, to enable the communication, different types of networking
switches and physical cabling infrastructure are implemented in data
centers. The need for two different kinds of physical network
infrastructure increases the overall cost and complexity of data center
operation.
Fibre Channel over Ethernet
(FCoE) protocol provides consolidation of LAN and SAN traffic over a
single physical interface infrastructure. FCoE helps organizations
address the challenges of having multiple discrete network
infrastructures. FCoE uses the Converged Enhanced Ethernet (CEE) link
(10 Gigabit Ethernet) to send FC frames over Ethernet.
