I'm preparing to head off to the Uptime Symposium 2010 that's being held from May 17th through the 20th in the New York Hilton with a great deal of anticipation. The central them of this event is Data Center Efficiency and Green Enterprise IT, topics that are parallel to system software, virtualization technology and cloud computing and so, of great interest. We're watching the world of IT and Facilities collide.
Why is an old software hack like me interested in a datacenter event? It's due to an epiphany I experienced at a Site Uptime Network meeting towards the end of 2009. After speaking with facilities managers from some of the largest, most important datacenters in the Southeastern U.S., I suddenly understood that there were worlds of technology and expertise that paralleled what I knew, but were totally separate.
I knew what was going on inside of the systems, storage and networking equipment. I could sketch out block diagrams of complex, multi-site, multi-system workloads and discuss the pros and cons of using different types of technology at each tier of a multi-tier application architecture. It wasn't until I spoke with the people attending the event that it occurred to me that I had no idea what it took to design, implement and operate a datacenter that housed my toys.
I spoke with people who understood power, cooling and the processes and procedures that created a special environment that was suitable for my toys. I had never given it a thought. I just took it as a given that there was enough power, that the room was kept cool enough and that special environment was always there.
After speaking with them for a time, it became very clear that these people were industrial magicians of a first order, but their line of magic, the disciplines they followed were strikingly different from those I worked so hard to understand.
The datacenter of the future is going to require all lines of industrial magic to work together, unlike what we see in many places today.
System architects are designing workloads that can move around to optimize service levels and handle outages. The datacenter folks will need to deal with an environment that is more dynamic than they've ever experienced.
Facilities people are dealing with power and cooling issues that may require racks or, perhaps, whole rows of computers to be shut down when not in use to conserve power and reduce heat production.
Working together, these two teams are going to change the face of computing.
I'm looking forward to speaking with these fantastic people this week and will post about my experiences.