Extreme Heat Leads to Data Center Outages and Threatens Hundreds of Millions of Dollars of IT Equipment.  Is Philadelphia at Risk?

Energy efficiency is a significant issue throughout the data center industry but sometimes there is an even more important issue in thermal management, keeping the lights on and protecting your equipment during extreme heat events.  Now that summer is here, the heat is on for data center operators.

Service interruptions from extreme heat can damage both your business and your customer relationships.  Potentially losing millions of dollars’ worth of IT infrastructure within a data center that is broiled in just a few minutes exposure to extreme temperatures might have seemed far-fetched a few years ago, but the potential for extreme heat risk is now all too real.

Let’s examine some heat events in the data center world from last season.  

With summer here, and the Southwest cooking under an expected three weeks of temperatures at 110(!!!) degrees or higher, it is a near certainty there will be fresh instances to add to this list.

Sacramento:  Last September a Twitter data center was knocked offline by extreme heat.  The social media giant reported internally "on September 5th, Twitter experienced the loss of its Sacramento (SMF) datacenter region due to extreme weather. The unprecedented event resulted in the total shutdown of physical equipment in SMF."   

Twitter is a sophisticated data center operator but in a whistleblower document Twitter’s former head of security Peiter Zatko reported. “Even a temporary but overlapping outage of a small number of datacenters would likely result in the service [Twitter] going offline for weeks, months, or permanently.”

London:  The extreme temperatures in July 2022 in the UK saw a high temperature of 104.4 Fahrenheit at London’s Heathrow Airport, shattering the all-time record by nearly three degrees.  Resulting data center outages in London included a Google Cloud data center knocked offline as well as service interruptions to Oracle Cloud.

Washington State:  In early late June and early July 2021 Washington State suffered an extraordinary heat wave which influenced data centers in the Quincy and Wenatchee areas along the Columbia River.   There are dozens of hyperscale and colocation data centers in this area. 

The heat was astounding.  An unprecedented 116 degrees in Quincy shattered the all-time high record of 109, while Wenatchee hit an astounding 113 degrees.  At that point energy efficiency didn’t matter.  The focus was on maintaining operability and protecting the hundreds of millions of dollars in IT infrastructure housed in these facilities.  Even after hitting those all-time records, highs remained in triple digits for a full week.  Even Seattle saw temps of 108 degrees.

Fortunately, the Quincy-Wenatchee area had mostly relatively new infrastructure and technologies.  But even with data center professionals in the area working diligently there were some reports of heat-related issues. 

Pennsylvania at Risk?  Philadelphia doesn’t seem like a place where extreme temperatures would threaten IT infrastructure and data center availability, but neither do the Pacific Northwest or London.   In fact, our state has seen a high temperature of 111 degrees while the record high temperature in Philadelphia is 107 degrees.   As data center professionals it is our responsibility to prepare for worst case scenarios, and what happened in Sacramento, London, and the Pacific Northwest should be on our collective radar screens. 

Steps to take: 

1. Evaluate all of your emergency preparedness plans.  If in the summer of 2023 our region has an extreme heat wave with high temps of 105 or more, what will you do?   Evaluate your plan, make improvements if called for, communicate to your organization what would be done in such a scenario, and practice its implementation.

2. Consider your facility’s thermal properties.  Is it a new data center with state-of-the-art cooling infrastructure or do you own one of the many legacy data centers in our area that are beyond their planned useful lifespan? 

3. After taking actions on steps 1 and 2, evaluate your service options for your primary production data center, as well as your disaster recovery and business continuity.   Are there other facilities in the market that you should consider utilizing to help you overcome the potential business damage of an extreme heat event.

Direct LTx is well-suited for helping you both weather a storm and extreme heat in our data center 51 miles west of Center City Philadelphia. In addition to being 84 feet above the 500-year flood plain the Direct LTx data center is built into the top of a hill, with below-ground data halls limiting solar heating.  

Thanks to those natural attributes of our location, our risk during an extreme heat event is quite low, though we still have an action plan for it.   If your risk is higher, upgrading your heat preparedness plan is not just a wise thing to do, it is a must.

Are you and your team trying to determine whether the cloud, colocation, owning and operating a data center, or some hybrid combination of those factors are best for your future?   Direct LTx just published 8 Key Questions for Every Philly IT Decision-Maker:  Will Your Data Center and Cloud Strategy Propel Your Organization to Success or Leave You Behind the Times in our Challenging Market?    For your copy of this white paper, click here.