This weekly feature from S&P Global Market Intelligence, in collaboration with internet-service monitoring company ThousandEyes, aims to give remote workers insights into internet service disruptions.
As people returned to work after the new year, there was a sharp spike in internet outages in the U.S. and worldwide.
Global outages at ISPs rose 72% worldwide and 42% in the U.S. during the week ended Jan. 15. U.S. outages accounted for 47% of all outages worldwide, according to data from ThousandEyes, a network-monitoring service owned by Cisco Systems Inc.
"The global outage downward trend, which we’ve observed over the past few weeks, ended this week," according to Angelique Medina, director of product marketing for ThousandEyes. "This increase in global outages appears to coincide with the end of the holiday period for both EMEA and APAC."
It is difficult to match specific events with the frequency of network outages, but the number of outages did drop worldwide during each of the past four weeks that corresponded with the end-of-year holiday season.
On Jan. 13, AT&T Inc. suffered a large-scale outage that affected customers in the U.S., U.K., Japan, Germany, Canada, Australia, India, and the Netherlands.
The problem appeared in portions of the AT&T network near Phoenix at 9:25 p.m. ET on Jan. 13, and it persisted for 23 minutes before being cleared at around 9:50 p.m. ET, according to ThousandEyes.
Though widespread, the impact did not compare to the days of downtime suffered by AT&T customers in Tennessee, Kentucky and Alabama following a bombing that damaged an AT&T switching station in Nashville on Dec. 25, 2020.
Microsoft Corp. suffered an outage on its networks that limited access to Microsoft online services and applications. It also hampered the ability of partners downstream to connect their own customers to those services as well.
The outage appeared in portions of Microsoft's network in Des Moines, Iowa, and lasted a total of 12 minutes, beginning around 12:15 a.m. ET on Jan. 13.
Three outages of 4 minutes each were spread across a 30-minute period from 12:15 a.m. ET to 12:45 a.m. ET, when the problem was resolved. The timing and uniform pattern of the 4-minute periods of downtime suggest the incident may have been part of an automated maintenance process, according to ThousandEyes.
Outages that occur during early morning or late evening hours, when regular maintenance efforts are often scheduled, have a much lower impact on end-users working from home than those that occur during their workday, Medina said.
Despite the sharp increase in the number of outages worldwide, the proportion of outages occurring during business hours dropped overall to 35%, compared to 42% the week before.