How long can you expect a data center to last? Pitt Turner, Uptime Institute Executive Director, discusses the practical life expectancy of a data center, and how the decisions companies make up front determine the useful lifetime of a data center facility.
In this video, Uptime Institute Executive Director Pitt Turner answers the question on a lot of data center owner and operator’s minds: Can you build a Tier III data center using modular or containerized, pre-fabricated data center components. The short answer: Yes.
In this new video, graduates of Uptime Institute’s Accredited Tier Designer curriculum discuss what they’ve learned from their experience, and the benefits of ATD accreditation to their organizations.
“As a professional in the data center business, especially where I come from, having a professional certification is a competitive advantage in the market. In fact, it is sometimes specifically mentioned in the RFP. Attending the ATD course has helped me tremendously in my career and organization,” said Adel Rizk, Engagement Partner at Edarat Group. “Having people from all different parts of the world, each bringing his own data center experience, makes the course very fruitful. Honestly, I carry the book from the ATD course everywhere I go.”
In this keynote panel discussion from Uptime Institute Symposium 2011, Christian Belady (Microsoft), Bob Cashner (Wachovia), Mike Wills (BMO Financial Group), and Matt Stansberry (Uptime Institute) discuss how IT organizations from different business verticals approach data center cost reduction. Belady represents a cloud computing provider, while Cashner and Wills represent large banking institutions. While these organizations couldn’t be more different, they do share motivation to be hyper-efficient and effective with data center resources.
“Financial institutions are very IT dependent,” said Wills during the session. “The data is so critical to a bank. If we lost customer data, our credibility would be gone, and in essence we’d start exiting the business. Managing that infrastructure at the right price point is extremely critical to us. If we’re not driving our costs down from an IT perspective, we’re creating a lack of profitability. If you take a look at what’s happening in banking, our investments in IT and the infrastructure is huge. We’re in the neighborhood of 40% of our reinvestment in IT. that’s what banking is all about — internet banking, telephone banking. You start looking at what’s happening in the capital markets, it’s all IT dependent. Everything is really relying on the infrastructure. Our job is to drive those costs down to the bare minimum, it’s the only way to move forward. There are parts of our business we’re not going to do as efficiently as a cloud provider. Running a payroll system? That can be outsourced.”
For the full session and other sessions full videos, check out Uptime Institute Symposium’s The Path Forward 2011. The Path Forward 2011 comprises the presentations, reports, case studies, and videos associated with the Uptime Institute Symposium 2011, The Disrupted Data Center: Cloud, Cost, Capacity and Carbon (May 9-12, 2011, Santa Clara, California). Although many of these materials are freely available, some items are reserved for Symposium delegates until August 1, 2011. Symposium attendees will be receive an email notification soon with a password to view the protected items.
In this short video from Uptime Institute Symposium 2011, Chris Malone (Thermal Technologies Architect at Google) discusses the key data center best practices Google uses to drive high energy efficiency, and how to adopt those best practices to improve your own data center’s PUE.
Fix cooling first, it’s the biggest term in your PUE overhead and offers the biggest opportunity for improvement. Do that by managing air flow — separating hot and cold streams. Raise the temperature of the air coming into the rack to make it easier to use economizers.
Second, optimize power distribution scheme by minimizing conversions and using efficient UPS solutions.
Third, measure and improve. Google is currently reporting a 1.16 PUE, and the organization got there by measuring and constantly trying to improve. The typical enterprise data center is likely at a PUE around 2.0, which offers lots of room for improvement.
“These best practices are so simple that everybody should employ them and see good results,” Malone said.