Welcome to our 22nd Data Center Executive Roundtable, a quarterly feature showcasing the insights of thought leaders on the state of the data center industry, and where it is headed. Our Fourth Quarter 2020 roundtable offers insights on four topics: Energy security in the wake of power outages in Texas and California, the state of data center metrics, timetables for the return of tours and conferences, and how data security challenges will look like in edge computing.
Here’s a look at our distinguished panel:
- Sean Farney, the Director, Marketing & Business Development – Data Center at Kohler Power Systems
- Jeff Klaus, General Manager of Intel Data Center Software Solutions
- Amber Caramella, the Chief Revenue Officer at Netrality Data Centers, and Advisory Council member at Infrastructure Masons
- Steve Dick, VP & General Manager of Emcor Enclosures.
The conversation is moderated by Rich Miller, the founder and editor of Data Center Frontier. Each day this week we will present a Q&A with these executives on one of our key topics. We begin with a look at our panel’s take on energy security in the wake of the recent energy crisis in Texas.
Data Center Frontier: In recent months we’ve seen disasters prompt rolling power outages in two of the largest data center markets, California and Texas. Are current approaches to data center infrastructure and operations sufficient? Or are there new strategies to consider in the face of energy security challenges? What are the new strategies?
Sean Farney, Kohler Power: Grid downtime in Texas was simply a failure of imagination. Here in Wisconsin, we’ve got a foot of snow on the ground and are just coming off of two weeks of subzero temperatures. We’ve had no interruption in utility feed because extreme winter weather was part of the Basis of Design during the engineering phase. The folks designing the Texas grid just didn’t ask “what if” vis-a-vis the realm of weather possibilities – they didn’t imagine severe winter weather and protections against it weren’t built in.
Fortunately, we data center operators are a paranoid bunch and ask “what if” ad nauseum. We are also contractually bound to remunerate customers for uptime SLA violations and our annual bonus reflects it. This is why I stored 250,000 gallons of diesel fuel underground while running a facility for Microsoft; the lights had to stay on.
And this is why despite an unimaginative grid design, you haven’t heard about data centers going dark in Texas. We should actually be celebrating the prescient professionals who already had a plan for grid down and rolled right over to generator power. Unfortunately, the state will have a Texas-sized Site Selection blackeye due to power reliability concerns for several years.
Amber Caramella, Infrastructure Masons: The inefficiencies in California and Texas are bringing the conversation of resiliency to the forefront. The most recent issues in Texas last month, where a significant amount of generation capacity went offline due to the extreme cold conditions, brought attention to many of these unforeseen operational vulnerabilities.
Data centers across the state of Texas experienced a range of challenges, including failing electrical and mechanical systems and issues with fuel delivery. Many stayed on generators for an extended period.
Netrality’s 1301 Fannin did not experience any utility outages and maintained 100% uptime for critical and essential services, per our design. Additionally, we opened our data centers to employees and their families for safe shelter.
In markets susceptible to outages or natural disasters, data centers should have resources available that are over and above normal safeguards of a hardened infrastructures. As owner and facility operator, Netrality evaluates designs and redundancies in connection with market risks – increasing fuel storage and make up water levels in markets like Houston, for example, to safeguard our facilities. We believe this approach significantly mitigates the common threats to facility infrastructure particularly as it relates to external threats beyond our control.
Approaches to data center design are changing as well, with standards requiring data centers to be designed to withstand temperature ranges within ASHRAE’s 20-year extreme minimums and maximums. Other new strategies include implementing equipment with monitoring tools at every level that can anticipate problems before they occur and physical design elements that can adapt to the toughest conditions. By leveraging power intelligence capabilities at the device level, data centers can improve energy efficiency and anticipate customer needs.
With the increased discussions around the power grids and generation, I believe there will be more technological advancements in power infrastructure to ensure these disasters do not repeat. We are currently seeing innovative advances from Telsa with a 100-megawatt storage project currently under construction outside of Houston.
Jeff Klaus, Intel: Power management is a challenge due to various uncontrollable factors including growth and sourcing, but there are ways to make better decisions and control operations in stressed situations.
These items include monitoring at a subsystem level, looking for optimizing workload across deployed systems (to ensure utilization), and implementing power controls or power limiting to ensure utilization across whitespace stays within acceptable levels, particularly if running on backup power.
Looking into how the IT demand can be shaped to help the data center as a whole remain functional is a strategy that we’ve seen deployed in future-oriented data center strategies.
NEXT: The state of data center metrics.
Keep pace with the fact-moving world of data centers and cloud computing by following us on Twitter and Facebook, connecting with DCF on LinkedIn, and signing up for our weekly newspaper using the form below: