Chapter 9: Putting Storage on the Network: A Detailed Discussion

Network Attached Storage (NAS) is the evolution of network servers. It was developed from efforts to configure, optimize, and operate a server directed only toward one thing: providing shared storage for users on a network. NAS is one of two pillars that make up storage networking infrastructures in todays data centers. It has evolved into a class of storage devices that address storage needs from departmental storage appliances to enterprise-class storage servers managing multiterabytes of data. This section will explain, discuss, and evaluate the NAS architecture, from the hardware and software required, to the necessary connectivity components . Developed using concepts from Part II, and expanding the NAS overview from Part I, this information will encompass a detailed view of the NAS internal structures.

The NAS Architecture

NAS is a specialized computer that provides file access and storage for a client/server network. Because it is a specialized solution, its major components are proprietary and optimized for one activity: shared file I/O within a network. NAS is considered a bundled solution consisting of prepackaged hardware and installed software regardless of vendor selection. The bundled solution encompasses a server part, a computer system with RAM, a CPU, buses, network and storage adapters, storage controllers, and disk storage. The software portion contains the operating system, file system, and network drivers. The last element is the network segment, which consists of the network interfaces and related connectivity components.

Plug and Play is an important characteristic of the NAS packaged solution. Given the bundling and configuration that comes with the product (which is dependent on the class of NAS solution you are implementing), its out of the box and attached to the network in less than five minutes. Although this may be a fairly accurate portrayal for installation and identification to the network, the Plug and Play value must be considered in context with the size and scope of the NAS device. There are many implementation caveats to consider and ample thought should be given to existing server clusters and storage systems already in place.

However, for most small networks, providing a Plug and Play NAS solution is a productive solution. Though more sophisticated departmental solutions will take longer, the bundled characteristics of the NAS devices save considerable time in planning and implementation. Of course, enterprise-level solutions should have much more thought put into their planning and implementation. Obviously, with this class of storage, care should be taken not to fall prey to the five-minute Plug and Play temptation which could result in much larger infrastructure problems.

NAS can be a productive solution. Its simplicity is one of its greatest strengths. The entire perspective of the NAS phenomenon , however, should be considered prior to any implementation. The following points summarize the issues that are important to know and understand to facilitate a long- term productive configuration:

• NAS is a specialized server solution with proprietary hardware and software.

• NAS is optimized for file I/O; support of any other types of workload I/O will prove problematic .

• NAS is one of he easiest ways to accommodate increases to storage capacity for LAN-based shared files.

• NAS can be one of the most difficult to administer and manage when numbers of boxes exceed reasonable complexities.

• NAS uses existing network resources by attaching to Ethernet-based TCP/IP network topologies.

• Although NAS uses existing network resources, care should be taken to design expansions to networks so increases in NAS storage traffic do not impact end- user transactional traffic.

• NAS uses a Real Time Operating System (RTOS) that is optimized through proprietary enhancements to the kernel. Unfortunately, you cant get to it, see it, or change it. Its a closed box.