In normal times, network availability is always at the forefront of business leaders’ minds. Any interruption in their connectivity takes services, apps and people offline that can cost as much as $400,000 per hour. Business as usual means remaining vigilant and trying to optimize networks for performance and reliability, while minimizing disruptions.
But these aren’t normal times.
In recent months, network resilience and availability has become exponentially more important as essential services and industries like medicine, education and government are pushed fully online. Meanwhile, both businesses and consumers are turning to virtual content platforms to stay connected and in touch, increasing network traffic as much as 40% in some areas and pushing many networks to their limits.
Increased global network demand is the new normal. And, as such, our industry must adapt to the new normal by rethinking its commitment to uptime and performance. In an age when industry standard availability remains at five-9s (99.999%) availability, CoreSite has pushed the envelope to achieve eight-9s (99.999999%) in 2019 to help to ensure that organizations, regardless of industry, have easy access to high-performance and reliable connections.
CoreSite’s ability to achieve eight-9s availability across nearly two dozen hyperconnected data centers is predicated on three factors: Plant, Processes and People.
The physical data center facility (“The Plant” to those of us in the industry), is the center of the connectivity universe. It’s where all the essential infrastructure — generators, cooling towers and chillers, electronics and emergency fire systems — resides.
Over time, every mechanical and electrical component of a data center will inevitably degrade. It’s natural and expected. But how frequently and how well each of those components is maintained can be the difference between an interruption that impacts customers and one that goes virtually unnoticed.
CoreSite’s approach to plant maintenance translates to nearly 150,000 maintenance hours—50/50 planned/unplanned—invested annually into maintaining every piece of equipment across each of our facilities in an effort to sustain high-performance and protect against future disruptions.
Each year, CoreSite reviews its maintenance and repair standard operating procedures (SOPs) to compare them against both industry standards and manufacturer recommendations. We actively adapt practices that support improving the reliability of our infrastructure, optimizing data center performance and extending the functional life of each component to help minimize capital costs.
Maintaining exceptional performance doesn’t happen without a coordinated effort across teams and departments. CoreSite’s operations team is guided by 7 Principles of Operational Excellence that underscore our commitment to consistent, repeatable, and effective processes for optimizing our data center operations.
Every operations employee trains on each principle and passes both theoretical and practical exams to ensure consistency across the organization. Risk assessment is a core element of operational excellence, which is why we continuously monitor risk logs from standard maintenance reviews, walk-throughs and customer visits as well as conduct periodic full site assessments. All risks — single points of failure, equipment aging, security metrics and fuel delivery timeline scenarios — are loaded into a database and graded on magnitude and likelihood to arrive at a risk score. The risk score helps prioritize particular maintenance activities and maintain focus where it belongs.
In addition to risk assessment, teams routinely gather to review every incident in the facility whether it had an impact on customer operations or not. They take a deep-dive into the event to understand which point in our chain of operations needs bolstering, what lessons we learned from the event and how to revamp processes to make those changes a reality.
The single biggest factor in achieving eight-9s uptime is having the right people running the business. CoreSite’s operations staff boasts 950 years of combined industry experience. Every aspect of the operation— from security and engineering to sales and executive leadership — is staffed by industry veterans from diverse backgrounds, unique skills and individual personalities who each contribute to producing the highest possible level of performance and the best possible customer experience.
Their contributions are enabled by an industry-leading and unique employee training initiative that features 49 individual programs spread over six tiers of qualification. The programs cover everything from electrical theory and mechanical design to building automation and energy efficiency assessments and require employees to pass both theoretical and practical exams to qualify for promotion.
This matrixed, cross-training approach helps to minimize knowledge gaps and ensure our entire team — no matter which office or facility they’re in — follows the same best practices and operational standards to ensure we’re doing everything possible to maximize network performance and availability.
As the world grapples with sudden and historic changes to daily life, network providers are under increasing pressure to maintain connection quality and speed even as everyone — businesses and private individuals alike — go online en masse.
CoreSite’s proactive approach to reaching and maintaining eight-9s uptime across our entire footprint may have seemed like overkill in previous normal times. But today, it’s proving vital for keeping everyone online with the high-performance and reliable connections they need to evolve and adapt in a rapidly changing global environment.
Contact a CoreSite solutions expert to learn more about how industry-leading network availability can help your business thrive in uncertain times.
SVP, Data Center Operations
Anthony Hatzenbuehler is CoreSite’s Senior Vice President of Data Center Operations. He is responsible for CoreSite’s onsite data center operations activities, client services, and the Operations Support Center (OSC).Read more from this author