Why does it take the average data center owner/operator 3-7 years to adopt a technology solution in their data center and why is the issue important to consider?
Today power utilization in US data centers comprises roughly 2-3% of the total used by the entire nation. There are lots of arguments about why this is important or why it might even point to reduced power utilization in other areas (I.e., improved efficiency of booking travel without driving to the travel agents office). However, these arguments really don’t matter. The fact remains that our data centers are a huge target of opportunity. This target isn’t just because our DCs use so much power, but the fact that the vast majority of our DCs have potential efficiency gains amounting to anywhere from 25% to 80%. If we take the average between 25 & 80 of 52.5% you can begin to recognize the size of the target, it’s HUGE!
How does the rate of new technology adoption play into this? Well, many of the solutions available today (Virtualization, Airside/Waterside AC, 480 Volts to the rack, Powering off Servers, Hot/Cold air isolation, etc., etc.) are proven and used in production environments. So, why aren’t we lapping up these technologies like a bunch of hungry junk yard dogs? Unfortunately, that’s where it gets a little more complex. I believe there are a number or reasons, like myth, risk aversion, lack of knowledge, higher priority projects, limited or poor integration between IT & Facilities teams, and a large number of data centers without dedicated resource with specific data center roles and responsibilities. We also don’t have a common certification program for data centers, nor do we have a simple way for folks to measure their performance, other than the PUE or DCiE metric created by Green Grid.
Myth: “I heard that doesn’t work” or “the risk to the environment is high”
Risk Aversion: The data center is the beating (mechanical & digital) heart of the modern company, and we’re trained not to mess with it!
Lack of Knowledge: Data Center staff (IT & Facilities) aren’t given the opportunity to learn more about what it takes to own a data center and own it efficiently.
Higher Priority Projects: While in many cases there are projects that are “higher priority”, I believe the reality is that IT & Facilities often don’t truly understand the potentially opportunity of improving efficiency in the DC.
Limited or Poor Integration between IT & Facilities: This issue helps lead to the problems noted in bullets 3 & 4. Because the two teams don’t work together they miss the changes that are currently affecting data centers. Unfortunately many of these “technology” changes have a facilities component and are therefore considered “Facilities” by IT and vice versa “IT” solutions which can improve facilities operations (like saving power) aren’t appreciated by IT.
Dedicated Data Center Roles & Responsibilities for IT & Facilities teams: This is generally caused by one or both of the following; small size of data center doesn’t seem to warrant dedicated resource and that’s combined with a limited understanding of what improvements can mean to the bottom line for the company.
Certification & Metrics: This is a major topic all by itself, so I’ll cover it in more detail in an upcoming blog. See my previous post for information on the effort I have underway to develop certification for data centers.
How do we solve these issues?
Our IT & Facility leaders need to work closely together to identify, vet and propose new solutions for the data center.
Our Leaders need to convey an expectation that risk can be OK. The Data Center manager, operator and or the facilities manager, engineer need to outline what the benefit potential is vs. the assumed risk of making the change. We can’t continue to say “it might cause a problem” and use that as an excuse to not implement.
We also need to take responsibility for getting through the FUD (Fear Uncertainty & Doubt) that many vendors put out there regarding other products. This is a common issue across IT & MEP (Mechanical Electrical & Plumbing) solutions.
What can the customer do as a group to help reduce the risk, accurately outline the improvement potential and demonstrate the efficiency gains in a meaningful way?
- Take advantage of groups like Data Center Pulse, Green Grid, 7/24, Data Center Dynamics and others to compare notes with other owner/operators.
- Develop knowledge of how and where new solutions are being used by other owners.
- Get involved in the solution. Don’t wait for the vendor to spell out the opportunity for you. If you do, you’re most certainly going to pay more for less.
- Develop roles and responsibilities that clearly define how your data center should be managed and maintained by IT and Facilities staff.
- Take a risk.. Just be smart about it.
Data Center Technology & Thinking is changing more quickly than ever:
It’s more important now than ever for owner’s to understand the recent and on-going technology changes and new solutions that are available for implementation in their data centers. Over the last 5 years we’ve experienced more change in what’s available for your data center than we did in the previous 15. If your data center is more than 2 years old there are most certainly major opportunities for you to take advantage of. As a group we have a common responsibility to our businesses and to the environment.
Act a lot, act a little, but please act.