When is a server not a server?
When it's a Storage Server.
Ask people what a storage server is, and you can expect to hear a variety of answers. Some will say it is a regular server with added features, a few describe it as a stripped-down box dedicated to a specialized function, and still others believe the term refers only to a network attached storage (NAS) box.
This article will attempt to define a storage server, differentiate it from a regular server, and give examples of storage servers on the market.
Not Your Average Server
The typical server is configured to perform multiple functions. It operates as a file, print, application database, Web, or miscellaneous server. As such, it must have fast chips, more RAM, and plenty of internal disk space to cope with whatever end users decide to do with it.
Storage Server Differentiators
Not so with a storage server. It is designed for a specific purpose, and thus configured differently. It may come with a little extra storage or a great deal.
"A general-purpose server typically has five or less disks inside," says Graham Lovell, senior director x64 servers at Sun Microsystems. "A storage server, on the other hand, has at least six, and more, usually 12 to 24 disks."
Storage servers are normally individual units. Sometimes they are built into a 4U rackmount. Alternatively, they can consist of two boxes — a storage unit and a server located near by. Both boxes can then be placed side-by-side in a rack. The Sun StorEdge 3120 storage unit and SunFire X4100 server, for example, can be combined into a storage server and placed in a rack.
Apart from extra disks, what else is different about storage servers? In many cases, they come with a host of specialized services. This can include storage management software, extra hardware for higher resilience, a range of RAID configurations and extra network connections to enable more users to be desktops to be connected to it.
Just a NAS Box?
Interestingly, some vendors define storage servers purely in terms of NAS. A NAS appliance (also known as a NAS filer) generally has a slimmed-down OS and file system, and only processes I/O requests by the main file sharing protocols. The big advantage of the NAS architecture is that it enables storage to be rapidly added by plugging the appliance into a network hub or switch.
"As far as HP is concerned, a storage server is NAS," says Jim Hankins, product marketing manager for HP's NAS division. "In essence, it is a dedicated file and print server."
HP has five ProLiant models available as general-purpose servers or storage servers/NAS filer — each has the same basic hardware configuration. If licensed as a storage server, the user may not run general-purpose applications on that server. If the same ProLiant server is being used as a regular server, however, applications can be run on it. To sweeten the deal, HP prices its storage servers a little lower than their general-purpose siblings.
NAS, it turns out, isn't really storage networking. Actual network-attached storage would be storage attached to a SAN. NAS, on the other hand, is just a specialized server attached to a local-area network.
In addition, HP's NAS-based storage servers have extra functionality built into the operating system — storage-specific management tools, "quota-ing" features, storage reporting capabilities, and a Web-based user interface that makes it easier to configure file and print. These features are not available on its general-purpose servers.
"For a little bit less, you get a lot more value add," says Hankins.
So is NAS really just a storage server? The answer varies, depending on whom you ask. But it appears there is very little difference between them. NAS, it turns out, isn't really storage networking. Actual network-attached storage would be storage attached to a storage-area network (SAN). NAS, on the other hand, is just a specialized server attached to a local-area network. All it does is make its files available to users and applications connected to that NAS box — much the same as a storage server.
"NAS is a marketing term," says Dan Tanner, an analyst at storage consulting firm ProgresSmart. "NAS is really nothing more than a file server, but specialized or adapted to the single purpose of serving files."
And what a marketing campaign it has been. From nowhere in the mid-1990s, Gartner projections predict the NAS market will exceed $2 billion by 2008, with an annual growth rate of 9 percent. And those numbers don't take into account a new NAS flavor called the NAS gateway. These gateways act as a file serving portal into a SAN: There are disk arrays in a Fibre Channel SAN that have a storage server on the perimeter acting as a NAS gateway. This is a one way to marry up NAS and SAN assets.
"There are two flavors of storage servers," says Hankins, "NAS appliances that have the disk storage in the appliance, and NAS gateways."
The ProLiant DL585 storage server is one example of a HP NAS gateway.
While some vendors use the same box as a plain vanilla server, others use a scaled-down version that is adequate for file serving. This is explained by Steve Duplessie, senior analyst at Enterprise Strategy Group. He defines a storage server as an optimized appliance designed to feed information, via a network, to a user or an application. As such, it is not typically compute heavy, but it has been designed from the ground up to provide specific I/O capabilities along with data protection capabilities.
"A regular server has to be generic, it doesn't know what kind of load demands it will have — gaming is much different than running a database, for example. A storage server, such as a NAS box, is a contained appliance that does one thing really well, like file serving.
What does a "regular" server have that a storage server doesn't? According to Duplessie, it typically has more processing power, more RAM, and a more generic I/O structure and file system. As a result, most storage servers perform at 50 percent of the performance of a regular server for the same function, he says.
What does a "regular" server have that a storage server doesn't — More processing power, more RAM, and a more generic I/O structure and file system.
This trend toward specialized computing elements is far from new. TCP/IP routing, for example, was a function that every operating system ran — until Cisco came out with a dedicated box that did it far better than hosting it on a general-purpose server.
"Any time you can optimize a function, it will be better [on a specialized box] than if executed on general-purpose gear," says Duplessie.
Dan Tanner, an analyst with the storage consulting firm ProgresSmart, agrees with Duplessie's view that a storage server is a specialized server or appliance.
"The server OS is cut down to address purely print server or file server functions, and often contains specially tuned or enhanced code," says Tanner. "Before NAS came along, though, Microsoft said you could use a regular server for file serving."
But using a vanilla server for file serving could lead to problems. Administering a general-purpose server is more complex. Further, someone might be tempted to use the server for multiple functions. Dedicated storage servers, therefore, have become the norm.
Not surprisingly, Microsoft introduced Windows Storage Server 2003 to distinguish it from general servers running the Windows 200x operating system. Windows Storage Server 2003 is a dedicated file and print server based on Windows Server 2003 and tailored to networked storage. It supports file serving and backup and replication of stored data. It can also be used to consolidate multiple file servers into a single box.
Windows Storage Server 2003 includes advanced availability features, such as point-in-time data copies, replication, and server clustering. They come in sizes ranging from a few hundred gigabytes to several terabytes. It is available in pre-configured NAS appliances from vendors such as HP and Dell. As a result, IDC reports NAS appliances running Windows now account for about half of all appliances in the market.
Storage Servers vs. Disk Arrays
Just as there is some confusion between ordinary servers and storage servers, there is also sometimes a misunderstanding between storage servers and disk arrays. Exactly where does one end and the other begin? A storage server can have as many as 24 disks — enough to quality as an array. Disk arrays, however, can have hundreds of disks. So where do you draw the line?
"A storage server is usually stand-alone and not connected to other servers," says Lovell. "Multiple servers, however, typically connect to a disk array."
Disk arrays, too, often connect to a server that could be styled a storage server. The storage server is the intelligence that goes in front of the array. In this arrangement, the server can manage several tiers of storage. It can even arrange the replication of data from one tier to another.
"A storage server serves the storage, and the disk array is the storage," says Tanner. "Using a storage server lets you use multiple or different arrays."
Duplessie further separates the two terms.
"A storage server typically speaks to files and talks to people or applications over Ethernet," says Duplessie. "A disk array is a low-level block device that only speaks to an operating system."