Storage network wars: Taking the best of both

Traditional server-attached storage implementations are being increasingly substituted by Storage Area Networks (SAN) and Network Attached Storage (NAS).

Though flexibility, quick and low-cost storage deployment favor of these technologies, SAN and NAS have always been perceived as competing technologies. However, the truth is that, they actually complement each other to provide access to different types of data. Each is designed for explicit environments and applications.

SANs are optimized for high-volume block-oriented data transfers while NAS is designed to provide data access at the file level. Both are based on open industry-standard network protocols — TCP/IP for NAS and Fibre channel for SANs. To expedite more flexible storage access, both technologies fulfill the need to remove direct storage-to-server connections.

The difference

SANs are networked infrastructures that provide a high-performance, highly scalable and flexible storage environment. This is accomplished by enabling direct connections between servers and storage devices such as disk storage systems and tape libraries.

NAS solutions are commonly configured as file-serving appliances. These are accessed by workstations and servers through a network protocol such as TCP/IP. Applications such as Network File System (NFS) or Common Internet File System (CIFS) for file access is used.

SANs use Fibre Channel switches and Fibre Channel network protocols ensuring device connections are both reliable and efficient. Connections such as these are based on either native Fibre Channel or SCSI (through a SCSI-to-Fibre Channel converter or gateway). One or more Fibre Channel switches provide the interconnectivity for the host servers and storage devices in a "SAN fabric" which is like a netted topology.

The majority of NAS connections remain between workstation clients and the NAS file-sharing facility. These connections rely on the underlying corporate network infrastructure to function properly.

SANs are ideal for mission-critical database applications, centralized storage backups, and high-availability and application fail-over environments. It offers scalable storage virtualization and improved disaster tolerance. This is because they are optimized to transfer large blocks of data between servers and storage devices.

NAS enables organizations to quickly and easily add file storage capacity to their technology infrastructure. Because it focuses specifically on serving files while hiding many of the details of the actual file system implementation, NAS appliances are often self-contained and easy to deploy.

NAS performance constraint is the ability of the network to deliver the data since network congestion directly affects its performance. NAS storage scalability is often limited to the size of the appliance enclosure. Thus, data backups in NAS environments are usually not centralized.

NAS is optimum for organizations that need to deliver file data to multiple clients over a network. These appliances also function well in environments where data must be transferred over very long distances. Being relatively easy to deploy, NAS appliances enable widespread distribution of NAS hosts, clients, and appliances throughout the enterprise. NAS provides reliable file-level data integrity. This is because file locking is handled by the appliance itself.

In conclusion, NAS and SAN are increasingly being used together. Numerous traditional differences between NAS and SANs are beginning to vanish. For example, NAS appliances increasingly use SANs to solve operational problems associated with storage expansion and data backup/restore.

However, NAS does not provide the full range of business benefits that SANs do. Nevertheless, compared to traditional server-attached storage, both SAN and NAS technologies reduce the total cost of ownership. They are the best bet for a good overall return on investment.