Once solution providers make the decision to build out their own cloud computing service and interesting shift in the perception of server vendors starts to take shape.
Most end customers don’t care what servers a cloud computing provider is using as long as the can maintain the quality of the service. That basically means that the brand of the server manufacturer is not as important as it once might have been. In fact, all things being equal when it comes to service and support, companies such as Super Micro now have a real shot at gaining ground on more established server vendors such as Hewlett-Packard, Dell and IBM.
For example, Super Micro recently rolled out a MicroCloud blade server architecture that allows eight hot-pluggable nodes in a 3U chassis. Each node supports a single Xeon e3-1200 processors, a PCIe slot and twin GbE ports. Future plans call for expansion up to 32 nodes for highly dense clusters of either low-power or high-performance machines. In an age of virtualization in the cloud, that kind of density is greatly appreciated by builders of cloud computing platforms. Traditional server vendors have yet to come up with anything comparable, says Don Clegg, vice president of marketing and worldwide business development for Super Micro.
In fact, Clegg says that one of the primary benefits of working with Super Micro is that cloud computing providers will generally get access to the latest processor technologies from Intel and Advanced Micro Devices before rivals that depend on HP, IBM or Dell. That’s could prove critical, says Clegg, when Intel, for example, starts shipping processors for servers based on its Sandy Bridge microarchitecture.
We all know by not that cloud computing represents a major shift in the IT delivery model. And if history is any guide, it’s usually the incumbents that have the most ti lose when such shifts occur.Tags: Xeon, virtualization, Super Micro, IBM, Dell, Hewlett-Packard, solution providers, channel, Advanced Micro Devices, Intel, Sandy Bridge, blade server, server, Cloud Computing