Government Computer News
Blade servers: Cutting edge
No matter how you slice it, blades offer advantages over conventional servers
Digg
De.licio.us
Slashdot
Technorati
In this report
Blade servers
The lowdown: Blade servers
What is it? A blade server is simply a chassis and one or more separate servers, each self-contained in its own card/enclosure. The chassis provides connections, power supply and management for all the blades.
Where should you use them? Blade servers are the best choice when there is either a large initial processing demand or capacity will be required to grow rapidly. Blade servers offer fast expansion with minimal effort. They are also a great choice when there is strictly limited space for new hardware.
The power supply, cooling and other support hardware in a blade server chassis are very robust; it has much more than is required for one or two blades. This isn’t a cost-effective alternative to a few standalone servers unless you are going to fill up most or all of the available blade slots fairly quickly (at least, well before the end of the useful lifetime of the chassis and original blades).
Must-Know Info Make absolutely certain that the management software is up to the task of managing a full load of blades and that it is very easy to use.
Remember, these tools must handle the equivalent of a room full of individual servers. Software that is good enough to manage a few individual servers might not be suitable for routine management of dozens of server blades, each perhaps with multiple processors.
The quality and ease of use of monitoring tools is especially important because you need to spot potential difficulties as quickly as possible. After all, having a single chassis fail is like having a dozen or more traditional servers crash all at the same instant.
Finally, although the cooling unit and power supply in a chassis look very impressive compared with those built into the single servers you are used to, you need to confirm they are really up to supporting a fully loaded chassis with the maximum number of processors and amount of memory and storage that can be loaded onto each blade.
Remember, too, blades must have dual power supplies, each capable of handling the entire load alone.
Consider the advantages that blade servers have over conventional servers:
- Easy expansion
- Lower cost
- Easier management
- Easier maintenance
- Lower power consumption
- Better utilization of physical space.
The actual servers, or blades, are plug-in cards containing one or more processors, memory, one or two hard drives and other components—everything, in fact, except the shared hardware supplied by the blade server chassis.
A complete blade server includes a minimum of one chassis and one blade, but the chassis is engineered to physically support a multiplicity of blades, each of which plugs into the same chassis.
For small operations, one or several individual servers are often the most cost-effective way to support a small network, but in larger installations, particularly ones that are expected to grow quickly in capacity, the blade server architecture is far superior.
Since the power supplies and other common components are not duplicated for each blade, this saves space and money. It can also increase reliability and decrease energy consumption—including a reduction in cooling load.
In addition to adding more blades as needed until the chassis is full, a user can often upgrade individual blades for many blade servers by adding more RAM, additional processors and larger hard-drive capacity.
Given all the options, it might be possible for a single blade-server chassis to be expanded to 20 or even 100 times the original processing capacity without taking up any additional space—simply upgrade blade capacity and increase the number of blades housed in the chassis.
![1105 Government Information Group [valor]](/images/1105logo_website.gif "Government Computer News [valor]")
© 1996-2008 1105 Media, Inc. All Rights Reserved.
