In a previous blog post I outlined why running an effective data center operations team involves many of the same elements as running a successful high-performance race car team. In this post I’m going to stick with the car analogy to explain the difference between a traditional facilities maintenance team and a critical facility operations team, such as those that handle data center operations.
Think of traditional facility maintenance, such as for an office building, as the local garage where you bring your car for routine repairs. Critical facility operations, on the other hand, is like a Formula 1 race car team. At least nine elements differentiate the two teams:
- Performance monitoring
- Emergency preparedness
- Financial risk
Professionalism, precision, and standards
Critical facility operations programs require a high level of professionalism in terms of mindset and attitude. Whereas a traditional facilities management program simply focuses on the basics – making sure the lights are on and the heating and cooling are working – a critical facilities operations program rigorously adheres to industry best practices. It’s like how your local mechanic can certainly change your oil and rotate tires according to manufacturer instructions, but not in a few seconds like the Formula 1 crew does.
It’s a similar situation with respect to precision and accuracy. In a critical facility, if various elements aren’t tuned precisely it could mean equipment failure and potential customer downtime, which is unacceptable. There’s lots more wiggle room in a non-critical facility.
Standards and regulations align to data center operations best practices – things you should do or, in the case of regulations, must do. Your mechanic likely refers to a list of routine maintenance that should be performed at certain mileage intervals, similar to the routine maintenance in a building. As you can imagine, the list is far more extensive and detailed for a Formula 1 vehicle, right down to how much it weighs and the octane of the fuel in the tank.
Monitoring, reporting, and preparedness
Performance monitoring in your local garage or typical facility is fairly limited. A mechanic may hook your car up to a machine that gives him some baseline diagnostics on how well different systems are functioning – whether your oil pressure is up to snuff and if your catalytic converter still works. Formula 1 and critical data center operations teams check on hundreds of elements and with a big differentiator: they do it in real time, and make necessary changes on the fly.
Documentation and reporting in the garage likely comes down to a repair manual that may be years old, and a list of services you’ve had performed in the past. Formula 1 and data center teams document every little tweak, including who made it and why. In a data center, detailed documentation and change management systems are crucial to preventing downtime, just as it’s critical to getting peak performance out of a race car.
In terms of emergency preparedness, if your car breaks down you make a call and wait an hour or so for a tow truck to bring the vehicle to your mechanic. On a race track, teams of professionals stand ready to address any problem, having practiced for just about any conceivable scenario. Critical data center operations teams likewise conduct emergency preparedness drills and have in place detailed emergency response plans to ensure all personnel are aware and competent.
Financial risk, safety, and security
In terms of financial risk, if your mechanic fails to fix whatever was wrong with your car, you can take it back and have him try again. It’s an inconvenience, but that’s about it. The same applies if the power goes out in an office building. Not good, but not the end of the world. But if a race car fails to work properly, there’s a huge financial penalty to pay in terms of lost prize money. The same goes for a data center, where downtime can cost enormous sums of money – an average of $8,850 per minute according to a Ponemon Institute study.
Safety is of course a concern for the local garage, both for the mechanic and customers, who generally aren’t supposed to be in the areas where mechanics are working. But the dangers facing Formula 1 or critical data center operations folks are far greater, whether from race car crashes or high-voltage switchgear that can likewise cause fatalities if not handled correctly.
Finally, the level of security applied to traditional facilities and critical ones is night and day. I just mentioned that customers aren’t supposed to be in areas where mechanics work, but I bet most of us have been, such as when the mechanic wants to show you something that’s worn out. Well you’d never get near a Formula 1 race car or pit crew. Similarly, data centers have strong physical security and lists of people who are authorized to work in various restricted areas. If you’re not authorized, you’re not getting in.
Learn essential elements of critical facility operations
Hopefully you grasp the point: critical facility operations such as those applied to data centers takes a completely different mindset and approach as compared to managing a traditional facility. I’ve seen it first-hand in the hundreds of data centers Schneider Electric operates, including some of the largest in the world, for customers across the globe.
We’ve captured some of what we’ve learned in a free eBook, the “12 Essential Elements of Data Center Facility Operations.” Download it now to learn more about what it takes to operate a data center that can’t go down.