When the Northern Virginia cluster of Amazon web servers went down, they took GrubHub, Canvas, Turnitin, Kaltura, and Zoom with them. These services make up a large part of student life here at Rowan, encompassing the majority of student dining, the main method of communication between professors and students, the host platform of most coursework, class syllabi, and information, as well as software students and faculty rely on to create and check that coursework.

When Amazon Web Services (AWS) began experiencing these disruptions, the world faced a shocking reality: the web platforms we rely on every day have a home. These physical servers are fallible, and Monday’s outage was a perfect example.

Today, Amazon has cornered over 31% of the cloud market, hosting a colossal quantity of the world’s data, which comprises not just files but websites, apps, and entire systems.

Companies can rent out space in these servers for much cheaper than if they were to host their own servers, which take up both space, energy, and oversight to run. It’s not unreasonable for a smaller company to operate its platforms via AWS.

Still, we’ve put all our eggs in one basket. The AWS outage encompassed just one cluster of servers in northern Virginia out of over 1.4 million servers across the globe, and took out services essential to Rowan’s function. Imagine what would happen in the case of a natural disaster or failures of infrastructure. What services might be impacted then?

This potential problem wasn’t on the mind of most Rowan students or faculty members, because at the end of the day, worrying about the what-ifs isn’t productive. Instead, we trust people who do understand these risks to make decisions that best benefit the members of our university community.

When Rowan University migrated from Blackboard to Canvas in 2020, and Rowan has been relying on GrubHub since at least 2019, you would think that the host server would have been part of the conversation when they chose to change systems. 

We have entire departments at this school dedicated to technology, academics, and student life. When considering digital infrastructure, the host company needs to be taken into account just as much as the user interface, speed, and support. 

That’s because when we diversify our technology and the companies that host them, we can more easily weather the storms of outages and service delays. Imagine if only Canvas crashed, but not GrubHub or Turnitin or Kaltura, or Zoom. 

Obviously, that isn’t possible because all those services are hosted in the same place by the same company, but there are comparable platforms that Rowan could opt to use. There are independently operated systems, alternative cloud platforms like Google or Microsoft, and even the choice to return to brick-and-mortar operations.

For example, the GrubHub outage resulted in pandemonium, as the entirety of the Student Center dining’s operations came to a screeching halt. This is in spite of the fact that the Marketplace is set up like a cafeteria or food court, and students should theoretically be able to queue, order, and pay for their food in person.

Instead, workers have received no training in operating like a traditional food vendor, and there is no alternative framework to charge students and categorize their orders in the event of in-person ordering being the only option.

Once again, Rowan placed all their eggs in one basket.

It’s not possible to anticipate all the problems that could befall our digital products and address them before they happen, but it’s the duty of those who decide on the programs we subscribe to to make informed decisions that are protected from failure. At the very least, there needs to be alternatives in place should this happen again.

For comments/questions about this story, DM us on Instagram @thewhitatrowan or email [email protected]