ISCSI: What is Internet Small Computer System Interface?

As greater numbers of large and midsized enterprises are faced with accessing/transmitting ever larger data streams over greater distances/multiple points where speed is essential, 10GigE has become the standard. Consequently, the choice of Fibre Channel over Ethernet (FCoE), Fibre Channel over IP (FCIP) or Internet Small Computer System Interface (ISCSI) connectivity becomes a fundamental choice based on a number of factors. In order to understand how traffic managers make these decisions, we should start with an explanation of ISCSI and how it works.

The Internet Small Computer System Interface, or iSCSI, is an IP-based protocol for transmitting data and linking data storage facilities. ISCSI works for transmitting data over Local Area Networks (LAN) and Wide Area Networks (WAN) as well as the Internet itself.

It works equally well as a Storage Area Network (SAN) protocol, thereby allowing enterprises to utilize data center storage arrays to consolidate their data. Historically, the improved data retrieval performance of SAN has allowed enterprises to operate as though they are utilizing local hard disk storage versus data center storage. The benefits of location-independent data storage and retrieval are fairly obvious in all cases.

Enterprises can increase performance with ISCSI as there are fewer layers of abstraction between volumes and the localized PC. In other words, ISCSI improves PC performance (and potentially lowers costs) in the enterprise by allowing large storage arrays to connect to client systems without the need for custom hardware or cabling. This makes it equally flexible for small businesses where many programs cannot run over shared networks.

How ISCSI – Internet Small Computer System Interface works:

When an application attempts to read from an ISCSI device, the SCSI read command is encapsulated inside an IP packet (along with encryption procedures if necessary). The IP packet is then routed just like any other IP packet on the network. Upon reaching its destination, the IP packet is decrypted (if it was encrypted before transmission), the encapsulation is stripped off, and the SCSI read command is interpreted by the ISCSI drive.

The SCSI commands are sent on to the SCSI controller, and from there to the SCSI storage device. Because ISCSI is bi-directional, SCSI write commands are handled the same way in order to return data in response to the original request.

Although many enterprise managers consider FCIP for their storage-to-network connectivity, it can only be used in conjunction with fibre channel technology. The fact that ISCSI can run over existing Ethernet networks is usually a deciding factor where the costs of enterprise-wide data transmission infrastructure changes necessary for fibre channel are cost prohibitive.

In addition, there are a large number of ISCSI-fluent technicians available in the marketplace, which often makes their salary ranges more competitive than FCIP or FCoE technicians. Complementing this broad based availability is the reality that a number of vendors, including Cisco, IBM, and Nishan among others have introduced ISCSI-based products (such as switches and routers).

It is important to remember that there are a number of true ISCSI products on the market, but just as many which are more of a hybrid system. These have definite usability implications for administrator front-ends for the various applications using distributed storage and technologies.