Tip Sheets

Tech outage illustrates ‘cascading failures’ of homogenous systems

Media Contact

Becka Bowyer

A widespread technology outage has highlighted dependence on software from a handful of providers. CrowdStrike is working to fix problems created for Windows users of its tools by a recent update. The following Cornell University experts are available for interviews.


Gregory Falco

Assistant professor in Cornell Engineering

Gregory Falco, assistant professor of engineering, is a cybersecurity expert.

Falco says: 

“The incident is a great example of the cascading failures that can occur given our relatively homogenous systems that comprise the backbone of IT infrastructure. We often think, ‘it would be so devastating if AWS goes down,’ but there are more than cloud service providers that make up the Internet. 

“Cybersecurity providers are part of this homogenous backbone of modern systems and are so core to how we operate that a glitch in their operations will have similar impacts to failures in systems that are household names. 

“The failure that occurred is not unique. The scale and the cascading consequences of the failure are what makes this notable. Cyber products are just software and software is soft... it's easy to make holes in them. Today is certainly a nightmare of a day for not only all the companies impacted, but global insurance providers who underwrite business continuity.”

Ken Birman

Professor of Computer Science

Ken Birman, professor of computer science, is an expert on big cloud computing companies.

 

Birman says:

“Widespread technology outages like the current Microsoft disruption have occurred many times over the past 25 years, but again and again we have seen the industry respond effectively – much as our country pulls together to respond to power outages after violent storms, bridge failures and other major calamities. 

“Obviously, these events sometimes trigger worry especially for people who don’t have a deep understanding of how today’s internet and cloud works. However, the underlying problems are always quickly identified and resolved, function is gradually restored and then the companies most directly involved find ways to restructure their infrastructures to be less vulnerable. This said, we do need to be conscious of the degree to which we depend on technology, and not take it for granted. Anything people build can break, and computer technologies are no exception.”

Cornell University has television, ISDN and dedicated Skype/Google+ Hangout studios available for media interviews.