What is a blade?

Print Friendly, PDF & Email

Traditional clusters are made from 1U-sized compute nodes that are linked together via a network. As explained earlier, this kind of system design suffers from issues with management, fault-tolerance, and power consumption. A blade server attempts to address these concerns by placing minimal compute nodes within a shared chassis.

The “blades” are nodes that contain a motherboard (CPU, memory, inter-node connection) as well as any disks. The chassis houses the power supply and fans. This kind of setup allows for greater density in that more processors and memory can be placed in a physically smaller area. In addition, there is less worry about power failure and there are fewer cables to manage. Current blade setups also allow for a switch to be in blade form as well, eliminating the need for 1U-sized hubs.

Given these benefits, it is easy to see that blade servers are expected to rapidly grow in the HPC market. It may only be a matter of time before traditional 1U-based clusters disappear in favor of blades. The Blade.org community provides a collaborative forum for advocating blades.