On May 31, 2018 from approximately 12:11am - 12:45am PDT, Amazon reported a significant Internet connectivity outage in their AWS US-East-2 region, which resulted in a complete loss of connectivity from this region to at least 10 other services including Salesforce, Rackspace, and Akamai. The impact of this outage was a unique illustration of how modern apps and micro-services depend on an intensive mesh of inter-service communication between public clouds, CDNs, hosting providers, SaaS platforms, and cloud-based payment API endpoints. With enterprises building so much dependency on communications that cross infrastructures, networks and services that they don't own or control, it's more important than ever to understand these dependencies so they can manage the digital experiences they're delivering to customers and employees. Here are the top five things businesses can learn from the AWS US-East-2 outage when it comes to building modern applications.

#1. Public cloud is becoming ubiquitous

More and more organizations are becoming dependent on public cloud providers to host critical elements of their customer-facing or workforce-enabling applications.

#2. Public cloud doesn't stand on its own

To fulfill their business processes, apps hosted in public cloud providers like AWS are constantly making service calls to CRMs, and payment gateways, pushing data to CDNs, and talking with services hosted in private, colocation and hosting provider data centers.

#3. One Internet outage can ripple widely

AWS US-East-2 suffered a 34 minute outage due to a Internet connectivity that affected one Internet address block, but this outage affected many dozens of critical inter-service communications.

Figure 1: Complex web of inter-service communications.

#4. Enterprises lack visibility into microservices inter-dependencies

This outage wasn't about AWS or any of the other providers "going down." Rather, it was a simple connectivity issue. Most enterprises are still under equipped to understand all the inter-dependencies their microservices and digital transformation initiatives depend on.

Figure 2: Correlated outage details showing impact to external services.

#5. Business risk due to unpredictable Internet behavior is growing

The more modern an app is, the more API calls are involved, and the more hosting environments, cloud providers and SaaS services are communicating across the Internet to power that app. This means that as enterprises continue to modernize their apps and digitally transform their business processes, exposure to Internet outages is only going to grow.

##

About the Author

Alex Henthorn-Iwane, VP Product Marketing Alex leads product marketing at ThousandEyes, and brings a perspective gained from working on innovative networking and analytics technologies since the early days of the commercial Internet.