Posted on June 11th, 2010There are three basic types of servers to choose from: Tower, Rack, and Blade servers. They are all feature unique benefits and limitations that need to be taken into consideration when choosing which type is right for you. Common considerations that need to be taken into consideration include overall cost, space, power, cooling, and scalability. Each type of server excels and lacks in different areas, so making the right decision is essential to preventing problems in the future.
Businesses have come to rely on three different types of servers: Tower, Rack, and Blade servers. IDC research firm discovered that currently 40% of small and mid-sized businesses in the United State are using Dell servers, almost 25% are using HP, and no other server vendor has been able to capture more than 5% of the market. Even with the market being dominated by vendors, the type of server being used varies by a far greater amount. All three types of servers were designed to fit a specific set of needs, and it is important to understand what each of them can offer you. Here is a look at all tower, rack, and blade servers and their unique benefits and limitations.
Tower Servers
Tower servers are the most basic, and often most recognizable, servers available. They have been considered the workhorse of the IT industry for years. Tower servers are stand-alone units that have all of the traditional server components including hard disks, motherboards and CPU’s. Their size and cost is fairly comparable to a similar desktop PC. Towers are often the best choice for smaller businesses or other environments where available space is less of a concern. They are often very powerful and often include additional bundled software programs. The reason that they are more powerful is because they can have more drives than the other options, additionally they are much easier to keep cool.
The largest drawback of using tower servers is the space that they require. Since towers are limited to wall and floor space, their scalability becomes limited. Aside from being the largest servers available, they also have an average higher operating cost than other options. This includes power, networking, and management personnel. They are also not the ideal solution if mobility is a concern.
Rack Servers
Rack servers are specially designed to be stored in racks, hence the name. This type of server initially became popular simply because of their space saving abilities, especially within datacenters. Most major racks are designed to adhere to the IEEE standard and are measured in “rack units” or “U’s”. The main motivation behind the design of rack servers is to provide vertical and horizontal scalability. Additionally, many other complimentary electronic devices are also designed to adhere to IEEE standards. This means that a rack can hold not only rack servers, but also SAN devices, rack consoles, power backup devices, and much more. This means that you can store everything in one place.
The largest drawback to using rack servers is that they can be very difficult to cool down because so many can fit into such a small space. They are often less powerful than tower servers when compared server to server, however when compared based on space (per square foot) they are fairly comparable.
Blade Servers
Blade servers are very similar to rack servers in many ways, especially in design. However, many components have been removed from them and instead are shared among all of the servers within an enclosure. This often includes cooling and ventilation components, power, networking and other interconnects. This allows blade enclosures to be more economical and extremely scalable. Not all blades are for computing, but rather provide storage, system management, or connectivity for the entire system. There are even blade servers with no disks, but are instead combined with SAN storage for further scalability. Blades often meet IEEE standards so that an entire blade system can be placed in a rack. While many of the blade server advantages can also be found in rack servers, the additional advantages include hot swapping, less cabling, redundancy, and diminished power and size requirements. Hot swapping is the ability to add, replace, or remove a unit without shutting down the entire chassis. This, along with the added redundancy provides excellent reliability.
The biggest drawback of blade servers stems from the advantages. There are a number of components that need to be added to the chassis in order for blade servers to run smooth, including additional storage (SAN) and cooling and ventilation components. Additionally, blade servers tend to be a bit pricier. Also, blade servers can require a lot of energy, which can create greater long-term costs.
All of these servers, Tower, Rack, and Blade, hold unique advantages and drawbacks that need to be taken into consideration. It is important to compare your options and make sure that the type of server that you choose will meet your needs.
Note: This is a Guest Post by Roko Nastic. He is a writer and editor at WebmasterFormat. He writes on various subjects interesting to webmasters like content management systems, website optimization and maintenance, blogging and similar. Visitors to WebmasterFormat can enjoy a lot of useful advices, latest news and read dedicated server reviews.