ISP, cloud, and UCaaS service providers are tracked by ThousandEyes, which provides Network World with weekly updates on the performance and Internet health check-ups of these three service providers.

To gauge how well businesses are being served by the internet, ISPs, cloud service providers, and conferencing services (aka unified communications-as-service) all supply reliable services.

These service providers are being monitored by ThousandEyes to see how they’re dealing with the problems they’re having. You may expect a weekly overview from Network World, which Network World will summarise here, detailing the most interesting developments in the delivery of these services.

Updated Oct. 31

From 354 to 278 global outages in all three categories, there was a 21% decrease in the number of outages. They fell from 154 to 147 in the United States, a drop of 5%.

From 254 to 197 outages, down 22% worldwide, and 112 to 109 outages, down 3% in the US, ISP outages dropped.

The number of cloud-provider network disruptions around the world (22) and in the United States (five) stayed the same.

The number of network disruptions in the United States grew from one to eight.

During the course of the week, there were two significant outages.

Hurricane Electric experienced an outage on Oct. 27 that affected customers and partners from around the world, including the United States, the United Kingdom, Russia, Brazil, France, Hong Kong, New Zealand, Finland, Japan, Germany, and Canada. Over the course of two hours and ten minutes, there were eight distinct episodes of the outage. At roughly 9:29 p.m. EDT, Hurricane Electric nodes in New York and Chicago, Illinois, were the focus of the first, which lasted around five minutes. The second one started around 15 minutes after the first one appeared to be over. Nodes in Ashburn, Virginia, went down five minutes into the outage. Nodes looked to be back online at 10 p.m. EDT, but about 10 minutes later, the nodes in Chicago experienced a brief outage. Nodes in New York and Chicago went down again fifteen minutes later. The nodes in those places appeared to recover many times, but then went back to being down for the next five times. Around 40 minutes into the disruption, the service returned to normal.

Customers and partners in the United States, France, the UK, Spain, Singapore, Hong Kong, Vietnam, Mexico and Chile were affected by a Rackspace outage on Oct. 25, which occurred on Oct. 25. Around 10:46 a.m. EDT, there was an outage lasting around 28 minutes that appeared to be centred on Rackspace nodes in Chicago, Illinois. A little over an hour later, at 11:15 a.m. Eastern Daylight Time (EDT), some of the nodes appeared to be recovering.

Updated Oct. 25

From 387 to 354, the number of worldwide outages decreased by 9% in the week preceding. The number of outages in the United States dropped from 185 to 154, a reduction of 17%.

From 281 to 254, the number of ISP outages worldwide decreased by 10%, while from 150 to 112, the number of outages in the US decreased by 25%.

Cloud provider outages fell from 30 to 22 worldwide, and from 10 to five in the United States.

When it came to network outages, there was only one that occurred in the United States. There were two global disruptions last week.

During the course of the week, there were two significant outages.

Customers and downstream partners in the United States, Australia, the United Arab Emirates, Thailand, Malaysia, and Canada were affected by Hurricane Electric’s outage on October 19. Starting at 1:51 a.m. EDT, Hurricane Electric nodes in Chicago, Illinois, experienced three outages over a period of 34 minutes. In Los Angeles, California, outage circumstances lasted about eight minutes after the first incidence appeared to clear up for about five minutes. Around 2:25 a.m. EDT, the outage was restored.

Several downstream partners and consumers were affected by a Level 3 Communications outage on October 21st. The outage lasted a total of 11 minutes, spread out between two separate incidents that occurred during a 25-minute span. At roughly 5:30 a.m. EDT, the first occurred and looked to be centred on Level 3 nodes in the Phoenix area. A few minutes later, the Phoenix nodes experienced outage conditions again. Around 5:55 a.m. EDT, the problem was resolved.

Updated on Oct. 18.

Across all three categories, global outages climbed from 352 to 387, a 10% rise, and in the United States, from 184 to 185.

Despite the rise from 270 to 281, the number of outages at ISPs fell from 163 to 150 in the United States.

Cloud-provider network outages grew from 18 to 30 globally and from 3 to 10 in the United States.

Global network disruptions for collaboration apps grew from one to two, while the number of outages in the United States remained the same at one.

During the course of the week, there were two significant outages. This disruption affected GTT Communications’ partners and consumers in countries including the United States, China, the United Kingdom and Ireland, Ireland and Japan on October 13th. GTT nodes in London and San Jose were originally affected by the outage, which lasted for 44 minutes. Later, San Jose nodes looked to be back online. EDT, the problem was resolved at approximately 11:55pm.

As of October 11, there was an outage on Microsoft’s network, which disrupted their downstream partners and services running on Microsoft environments. First spotted at 5:05 p.m. EDT, the 34-minute disruption affected Microsoft servers in Amsterdam, Netherlands. Five minutes later, Frankfurt, Germany, nodes began experiencing outage conditions that affected more partners. The Frankfurt nodes appeared to recover at 5:15 p.m. EDT. The issue was resolved around 5:40 p.m. EDT. In light of how long it lasted and where the nodes were located, it is most likely a maintenance exercise.

Updated Oct. 11

The number of outages in all three categories grew from 323 to 352 in the last week, a 9 percent rise. There was a 14% increase in outages in the United States.

Across the globe, ISP outages rose from 239 to 270, an increase of 13 percent; in the United States, ISP failures rose from 132 to 163.

A 31% decrease in cloud provider network disruptions was observed. They dropped from six to three in the United States.

In the US, there was a single outage of a collaboration-app network, bringing the total number of outages to six, down from six in the previous week.

It was on Oct. 4 that Facebook’s backbone network suffered an outage, causing Facebook, Instagram and WhatsApp to be unavailable for more than seven hours to everyone around the world. Second problem: Facebook’s authoritative name servers stopped advertising routes to its servers because of the “unhealthy” backbone network breakdown about 11:40am EDT. The services were unavailable. The outage was traced back to an erroneous configuration modification, which Facebook has since fixed. The problem was resolved around 6:45 p.m. EDT, when connectivity to services was restored. Here’s a more in-depth look at what caused the outage.

When Telia Carrier, a leading Tier 1 Internet service provider (ISP), went down on Oct. 7, customers and downstream partners in more than a dozen countries around the world lost access to Telia Carrier’s services. Nodes in Ashburn, Virginia, and Stockholm, Sweden, where the outage was first noticed at 12:00 EDT, are no longer affected. Nodes in Newark, New Jersey, and London, England began exhibiting outage conditions five minutes after the ones in Sweden appeared to have recovered, expanding the number of nations and downstream partners affected. All but the London nodes appeared to recover within 35 minutes of their initial detection. After an hour and 14 minutes, the disruption was resolved at 1:15 p.m. EDT.

Updated on October 4th, 2018

Last week, the total number of global outages fell by 13%, from 367 to 323, compared to the week before. In the United States, there were 161 outages, which stayed constant.

Internet service provider outages rose from 233 to 239 worldwide, while in the United States, they jumped from 115 to 132, or 15%.

Global cloud provider network outages decreased from 41 to 26, a 37% decrease, whereas in the US they rose from four to six.

The number of network outages caused by collaborative apps has dropped from eight to six, globally and in the United States.

During the course of the week, there were two significant outages.

Using Oracle Cloud services in the United States was disrupted on September 30th due to an outage suffered by Oracle. When the outage was first noticed at 4:01 p.m. Eastern Standard Time, it looked to be centred on Oracle nodes in Frankfurt, Germany, as well as Arlington and Ashburn, Virginia. Only the nodes in Arlington remained down five minutes later, and it looked that they had recovered. Outage lasted seven minutes and was resolved at 4:10 p.m. EDT.

On September 29, NTT America experienced an outage that affected some customers and downstream partners across the US, Japan, and Hong Kong. A 25-minute span was required to accommodate two separate incidents, each lasting about 18 minutes. New York City’s NTT nodes were the focus of the first sighting, which occurred at 7:10 p.m. EDT on March 1. It only lasted for around four minutes before clearing up at 7:15 p.m. Eastern Time. Exactly five minutes after the first event, which lasted 14 minutes, the New York and Ashburn, Virginia nodes were hit again. Approximately 10 minutes after the outage in New York appeared to be resolved, and the issue in Ashburn was resolved around 7:35 EDT, the service was restored to normal.

Updated on September 27th, 2018

Last week’s global outages jumped from 276 to 367, a 33% rise over the previous week. 116 outages in the United States grew to 161, a 39% increase.

It is estimated that the number of global ISP outages has increased by 24% since 2010, and that the number of outages has climbed by 32% in the United States since the same year.

From 24 to 41 outages, the number of cloud-provider outages climbed by 71%, while in the United States, they rose from three to four.

The number of network disruptions for collaboration apps climbed by eight throughout the world and eight in the United States.

The number of network interruptions caused by collaboration apps reached its greatest level this year. Globally, they grew from two to 10 and from one to nine in the United States, respectively.

This week saw two significant disruptions.

This interruption affected consumers and downstream partners in the United States, Germany; Finland; Singapore; Hong Kong; Australia; and Thailand on September 21. Around 10:25am EDT, there was an outage on Oracle servers in Frankfurt, Germany, which appeared to be the cause. Later, some of the nodes appeared to have recovered, reducing the afflicted countries to Finland and Hong Kong. At roughly 10:50 a.m. EDT, the disruption was resolved.

An outage at Comcast Communications affected downstream partners and customers in the United States, Switzerland, China and Hong Kong on September 20, 2014. The 18-minute film.

You must read this article on how to get a custom email address.

Wahaab Siddique
Mr. Wahaab Siddique

Wahaab Siddique is the blogger and founder of, along with this, he is a student of Civil Engineering.