IBM Tape Solutions for Storage Area Networks and FICON

Chapter 1. Storage area networks 5

Several vendors, including IBM, offer a variety of network storage solutions. These currently

fall into two categories:

򐂰 File servers

򐂰 Backup-archive servers

NAS, therefore, offers a number of benefits, which address some of the limitations of parallel

SCSI. However, by moving storage transactions, such as disk accesses, and tasks, such as

backup and recovery of files, to the LAN, conflicts can occur with end-user traffic in the

network. LANs are tuned to favor short burst transmissions for rapid response to messaging

requests, rather than large continuous data transmissions. Significant overhead can be

imposed to move large blocks of data over the LAN, due to the small packet size used by

messaging protocols.

For instance, the maximum packet size for Ethernet is about 1500 bytes. A 10 MB file has to

be segmented into more than 7000 individual packets (each sent separately by the LAN

access method and then reassembled), if it is to be read from an NAS device. Therefore, an

NAS solution is best suited to handle cross-platform direct access applications, not to deal

with applications requiring large data transfers.

NAS solutions are relatively low cost. They are also straightforward to implement, since they

can fit into the existing LAN environment, which is a mature technology. However, the LAN

must have plenty of spare capacity to justify NAS implementations, or as is often the case, a

dedicated, parallel LAN is used for NAS traffic.

LANs today typically run at 10 MB/sec or 100 MB/sec. But the relatively new Gigabit Ethernet

is becoming more common. This requires a change of infrastructure from copper cables to

fiber-optic cables. Even when the bandwidth of SAN and LAN are equivalent, as in the case of

Gigabit Ethernet, a LAN is still best suited to message traffic and a SAN to storage data

transfers.

NAS typically operates at a file level, with a user attached to an NAS appliance (usually a file

server) requests a specific file. This can be a problem in the case of large files such as

databases.

Storage works in blocks and has no knowledge of files. Blocks map to groups of sectors on

the storage media where the data is actually stored. The file table that maps files to blocks

resides in the server and is maintained by the server’s operating system, and not in the

storage device. The block table which maps blocks to sectors resides in the storage device.

SCSI commands, which are used to read and write data to and from the storage devices,

contain block information, not file information. Why is all this relevant? It is because of an the

emerging iSCSI implementation.

iSCSI enables SCSI commands to be sent over IP networks, hence the name iSCSI. This

means that applications connected to an NAS can request smaller amounts of data by issuing

SCSI commands. At first glance, this may appear to be a panacea for NAS and a death for

SANs. However, it is still subject to the overhead of LANs, and most applications work on files.

You can use NAS separately or together with a SAN, since the technologies are

complementary. In general terms, NAS offers lower cost solutions, is quick and easy to install,

and makes file sharing simpler, but with lower performance and less scalability than Fibre

Channel (FC) SANs.

1.2.2 Storage area network

A SAN is a specialized, high-speed, intelligent, and well-designed network attaching servers

and storage devices. A SAN allows an any-to-any connection across the storage network,