Processing power demands continue to increase with no sign of abatement. Customers want Web sites available 24/7, users want email permanently available, and they're impatient when those requirements aren't met. Server rooms and data centers must host ever more hardware to meet on-demand society needs -- and blade server technology is coming into play.
Sometimes these requirements are temporary or transient in nature; a project might require a suite of servers for a fixed period of time. Procuring new hardware, configuring it, supporting it during the project's timeframe and decommissioning it at the end can create significant businesses problems involving capital expenditure, data center access and change control.
One useful way to circumvent this problem is to use blade servers. A shelf of blade servers will use power more efficiently than similar numbers of physical "pizza box" (1U) servers, utilize fewer network ports and require less rack space.
Blade servers combined with virtualization technologies -- such as VMware Infrastructure using ESX Server and System Center -- scale out the number of server instances that can be run on a pair of processors, and scale up server power by being able to use a single processor at first and add processors as business volume increases. There is nothing worse than seeing a server sit idle because the anticipated level of business didn't materialize -- except seeing servers collapse under the sheer number of users because the service was greater than anticipated.
But blade server technologies don't need virtualization to leverage potential efficiencies. On their own they are every bit as useful and scalable. Blade servers borrows their operating system design from the storage area network (SAN), inevitably found in every data center. Taking disks out of blades reduces power requirements a little more and makes the blade a commodity item. Should a blade fail, ownership of the LUN on which the operating system resides is simply switched, automatically or manually, to a spare blade. If you wanted to maintain a spare 1U traditional server in a rack with traditional storage, you would have to enter the server room, power down the server, remove local disks and power up the spare server.
However you look at it, blades provide scalable computing power rapidly, so you can deploy new servers to your customers within minutes rather than hours or days.
About the author: MCSE+M, Microsoft MVP, is a technical architect for Posetiv Ltd., a U.K.-based storage integrator. He is responsible for the design of Microsoft Exchange and other Microsoft Server solutions for Posetiv's client base in terms of the SAN and NAS storage on which those technologies reside. Arnold has been a Microsoft MVP in the Exchange discipline since 2001, contributes to the Microsoft U.K. "Industry Insiders" TechNet program and can be found in the Exchange newsgroups and other Exchange forums. You can contact him at mark.arnold@msexchange.me.uk.
