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.

Because data blocks are the raw material from which files are formed, NAS also has a block component. These blocks are addressed on a per-file basis using metadata (directory information) to define the file-block relationship. The block access methods of a NAS device, however, are typically hidden in the NAS enclosure. To the outside world, the NAS device is a server of files and directories.

NAS accomplishes the central goal of storage networking: the sharing of storage resources through the separation of servers and storage over a common network. Like SAN-based storage, NAS overcomes the limitations of parallel SCSI and enables a more flexible deployment of shared storage. In redundant configurations, NAS can also provide highly available, nondisruptive storage access.

As shown in Figure 2-9, a NAS architecture includes disk arrays for data placement, a NAS processor, and an external interface to the user network. This architecture has a number of implications. Whereas the block data transport of a SAN normally occurs over a dedicated storage network (or virtual LANs within a Gigabit Ethernet network), the file transport of a NAS device assumes direct connectivity to the end-user network. Thus, NAS performance is determined in part by the bandwidth available on the messaging network. In addition, although the NAS processor is optimized for file transport, it is essentially a thin server sitting on a LAN and acting as a front end to storage arrays. To differentiate NAS from a common network-attached file server, vendors of NAS products have attempted to make the thin server component so thin that it is invisible from the user perspective. The term "NAS" is therefore always linked in marketing literature to the concept of "appliance" a device that is simply plugged in to the network and requires little administration. Finally, the connection between the NAS intelligence and its disks is immaterial from a user perspective. Although SAN storage is predicated on high-performance gigabit interfaces for storage (either directly or via bridge products), a NAS device can rely on parallel SCSI, IDE, or Fibre Channel for storage connectivity.

Figure 2-9. A NAS appliance can use any type of physical storage

The challenge of NAS vendors is to make their products more appliance-like and to reduce the overhead of NFS/CIFS protocols over TCP/IP. To deal with latency, some NAS vendors support the Distributed File Access System (DAFS) over Virtual Interface (VI). DAFS relies on remote direct memory access (RDMA) techniques to move file data directly to systems memory. If a NAS controller can efficiently place file data directly into the memory of the client, transport latency is offset by the higher performance of file data placement.

Another trend in NAS technology is the separation of storage from the NAS processor or head, so that storage capacity can be more easily expanded using SAN techniques or shared with SAN-attached servers. As shown in Figure 2-10, a NAS head still performs its traditional role of file serving, but its storage is no longer hidden within the NAS enclosure. This lets you carve out part of the storage capacity for file requests while leaving the remainder available for block access.

Figure 2-10. A NAS head with shared SAN storage

The top end of NAS product offerings have used Fibre Channel arbitrated-loop disk drives for block storage. This approach results in higher performance and scalable storage capacity, but it presents vulnerabilities in the event of loop failure. Today, embedded-loop switch technology is used on the backplane of some NAS storage arrays to bypass failures and maintain stable operation. This technology was initially developed by Vixel Corporation and adopted by Network Appliance for its NAS filers.

At the low end of NAS products, vendors have introduced economical NAS appliances using internal IDE disk drives. The basic functionality of NFS or CIFS file sharing is maintained, although these systems tend to be limited in total capacity and give somewhat lower performance compared with the enterprise-class NAS products.