Uptime Institute Market Intel: Expect Cascading Failures to Increase
by Josh Anderson
NORTH CHARLESTON, SC — The data center industry faces a lot of challenges, and one that is particularly interspersed among them is downtime. So CapRE invited the Uptime Institute’s Scott Macintire to our Southeast Data Center Summit Spring 2018 Update, to provide findings from a Market Intel Report based on thousands of survey responses from across the industry. It was no surprise that he devoted a large chunk of time to this topic. Below, we highlight some of the topics that Macintire covered during his address, all of which zero-in on how organizations can make sure that mission critical assets keep running on all cylinders – and how that impacts your bottom line.
Cascading Failures Set to Increase
“We expect outages to create cascading failures across multiple sites and multiple services,” remarked Macintire. “We expect to see more issues. This is just because we are adding more complexity to the challenge. One of the recent outages that we saw, it was an enterprise end-user who was using a colocation provider to house their work sites. They were storing things in the Public Cloud on S3. They had a Direct Connect over to Amazon S3. There was a Nor’easter Cyclone in Virginia. I’m not even sure what a Nor’easter Cyclone is, but it sounds vicious. It took out the power to the Direct Connect, that went to that data center where they had their storage.”
“However, the application had to have that storage to execute,” continued Macintire. “And they only had one of these Direct Connect over to that site, so they were down for four hours. The service provider on that direct connect doesn’t have an SLA around those direct connects, because it’s kind of a secondary concern. And that had caused the company to look at how they’re architecting their applications for the cloud and for colocation. So now they’ve made it from just one direct connect to two direct connects, and hopefully it doesn’t happen again.”
Hybridization = Resiliency
Macintire then shared how a similar topic has to do with workloads. “63% of organizations say that having workloads across the board makes them resilient. I agree with this, you’ve got your primary D/R, you’ve got something out on public cloud, you’ve got something on-prem, that’s fantastic,” he mused. “But the challenges and issues come back to the networking, and architecting for that network. That’s usually the short goal or intent. 9% say that they’re less resilient. I’d love to talk to those folks though because I bet they’re people that aren’t really embracing this technology.”
Who’s Holding the Bag When it Crashes?
According to Macintire, only 44% of organizations actually have someone that’s going to take the hit when something goes down. “I think this number will go up. It’s got to go up,” he predicted. But who are the people that organizations are reporting as responsible? “That poor CIO,” he answered. “That CIO, who’s got the responsibility for data breaches already. They’ve got the cyber-security bag over their head. They’ve always had this outage over their head, but now it’s tied to revenue. Increasingly, it’s tied to revenue. So CIOs hopefully will get more pay and bigger teams, but that’s very quickly becoming a stressful job.”
Be sure to check out previous CapRE Insider Reports covering Macintire’s remarks:
- UpTime’s Scott Macintire: Cap and Grow Strategy Leading to Potential “Silver Tsunami”
- Uptime’s Scott Macintire: Data Center Industry Must Address Recruitment and Staffing Challenges to Keep Up With Growth
- Uptime’s Scott Macintire: The Future of Data Center Skillsets is All About Networks
- Uptime Institute Market Intel: Are You Ready for the Edge?
- Uptime Institute Market Intel: Only 28% of Organizations Calculate the Cost of an Outage
Scott Macintire is the Director for Client Engagement at the Uptime Institute, which is focused on improving the performance, efficiency, and reliability of business critical infrastructure through innovation, collaboration, and independent certifications. It is best known for its widely adopted “Tier Standard” and the associated certification of data center compliance. Scott’s career in information technology and Infrastructure spans 10 years and includes time building solutions for infrastructure providers and providing consulting support for IT transformation initiatives. Currently at the Uptime Institute, Scott is focused on deepening existing relationships and building new partnerships in the US as the Director of Customer Success. He is a graduate of Tulane University and spent his early career as an Infantry Officer in the Marine Corps.