I’ve been in some interesting discussions lately about how we design and manage our infrastructures. There’s one school of thought that wants to build in structure and control at every conceivable point in the infrastructure so that it can be monitored, measured and adjusted. The other school of thought looks to a design a flexible infrastructure that can support most applications without micromanaging them, we call this the cloud.
Structure and control are hallmarks of the old purpose built infrastructures we’ve been building for decades. In the past this was required because the equipment we’ve had was barely up to the task. Over the last decade we’ve seen the equipment get ahead of the performance requirements of most applications. Leveraging this excess capacity so we don’t have to micromanage our infrastructure is the best way to reduce complexity and cost. There are some applications that still need to be carefully designed, measured and monitored because of their mission critical nature, or stringent performance requirements. But let’s not add the complexity and cost where it’s not needed.
As always these views are my own, and do not necessarily represent the views of HP.
Follow me on Twitter. http://twitter.com/bladeguy
