Selecting the right server for the enterprise is a serious and costly consideration. Server resources get a great...
deal of attention, yet the server form factor is usually an afterthought, despite it being a necessary part of the evaluation process.
Organizations typically consider rack and blade server form factors. Rack servers resemble wide, flat, deep pizza boxes that are stacked and bolted into large, free-standing metal frames. Traditional rack server form factors are 19 inches to 23 inches wide, a multiple of 1.75 inches tall -- known as a rack unit or U -- and are about as deep as they are wide. For example, a 19-inch 2U server is about 3.5 inches tall.
Heterogeneous data center environments use rack servers where varied makes, models and vintages of hardware share a common physical deployment scheme. Each rack server has various power connections and network cabling options. In addition, rack servers provide indicators -- or a control interface -- that enable administrators to directly configure and monitor the server status. Larger rack server sizes -- such as 2U or even 3U systems -- can accommodate additional processors, memory and local storage disks.
But rack servers demand a high level of physical intervention on the part of the IT staff. Mounting the servers to the rack, connecting cables, and directly configuring or monitoring a disparate array of servers requires staff support and planning; all of these traditional activities are under scrutiny as organizations seek to reduce costs and the time associated with these tasks either internally or through a managed service provider.
When to consider a blade server form factor
Blade servers' popularity climbed around the same time as the push toward converged and hyper-converged infrastructures (HCI) began. The goal was to shift from a heterogeneous data center using multiple vendors for hardware and software toward a homogeneous, single-vendor IT environment that conveniently combines servers, storage and networking devices into a single system. HCI devices adopt a modular design where modules -- or blades -- slide easily into a dedicated frame with slots and predefined connectivity located on the frame's common backplane.
The net result is a faster and easier physical deployment of blade servers and devices, such as network switches and storage blades. Because blade systems are integrated, single-vendor platforms, a single management plane handles configuration and management.
Blade servers and devices could create vendor lock-in, though this is offset by the speed, convenience and benefits of single-vendor support. Blade servers pack as much computing power as a comparable 1U rack server, but they lack the physical space to include any significant amount of on-blade disk storage.
Fortunately, the choice between the rack and blade server form factors is not an all-or-nothing decision. Admins deploy blade systems to secondary data centers, remote offices or in established data centers for task-specific computing, such as a fledgling private cloud.
Although blade and rack systems are not physically interchangeable, these different server types can simultaneously exist in the data center.
Dig Deeper on Server hardware strategy
Related Q&A from Stephen J. Bigelow
Full virtualization and paravirtualization both enable hardware resource abstraction, but the two technologies differ when it comes to isolation ... Continue Reading
Organizations can cap their hyper-converged infrastructure costs when they deploy the Azure Stack HCI platform, but once they plug into the cloud, ... Continue Reading
You can implement ESXi on ARM -- or other RISC processors -- in micro and nano data centers. A nano data center is more specialized but also more ... Continue Reading
Have a question for an expert?
Please add a title for your question
Get answers from a TechTarget expert on whatever's puzzling you.