- Google Cloud Platform went down for just over an hour on Thursday, after it suffered a "significant router failure" in one of its data centers in Atlanta, which led to network congestion, according to the company's service status page.
- The outage was unrelated to traffic levels brought on by increased traffic amid the novel coronavirus pandemic, according to Urs Hölzle, SVP of technical infrastructure at Google. "Our network is not stressed by Covid-19," Hölzle said on Twitter.
- A separate service disruption began affecting Google Cloud infrastructure components on Thursday afternoon (PT). Services such as Dataflow, BigQuery and Compute Engine presented elevated error rates until Friday morning, according to the company.
Switching most human interaction over to the internet represents a real stress test of network infrastructure.
Though Google Cloud's outage was unrelated to the pandemic, soaring online traffic will also test the resiliency of cloud providers, as digital services adjust and adapt to record surges in usage.
Companies are advised to run fire drills to prepare for a scenario where, as more people work and study remotely, digital services face peak traffic.
In the U.S., 44% of the top 200 cities by population suffered some degree of network degradation last week, according to stats from BroadbandNow.com. In New York City, the current epicenter of the pandemic, download speeds fell by 24%.
The interconnectedness of the internet could spell trouble for the new normal, as one service's outage can lead to another site or platform going offline right as businesses highly depend on the web to sustain their operations.