Network sniffers find COVID-19 did not break the internet – though it was behind a massive jump in outages

Disruptions leapt 63% as lockdowns came into force after February

8 Reg comments Got Tips?

Global internet disruptions went up 63 per cent after February and remained elevated throughout the first half of 2020 compared to pre-pandemic levels, according to net and cloud researcher ThousandEyes.

In March, as much of the world entered pandemic-related lockdowns, the number of outages across ISPs, cloud providers, content delivery networks, and domain name systems hit around 1,200 globally, the report said.

Under normal conditions, the number of outages ranges from 700 to 800 per month. By June, they had fallen back but not to pre-pandemic levels: there were still 44 per cent more than in January.

Despite the increase, it was far from the disaster some feared, claimed Angelique Medina, product marketing director at Cisco-owned ThousandEyes.

"Despite early fear and speculation immediately following pandemic-related lockdowns, the state of the Internet was and remains healthy, with our network measurements showing little evidence of systemic network duress, even when traffic shifts and volumes were at their peak. With few exceptions, Internet-related infrastructures over the last six months have held up well."

While the outages were a result of changes in behaviour during the pandemic, they did not directly stem from the internet being under "duress", the report argued.

"Large-scale network measurements collected between January and July 2020 suggest that recent increases in disruptions are not due to network duress, but instead the result of operators adjusting their networks to accommodate changes in traffic pattern and load," it said.

ThousandEyes gathers network metrics such as loss, latency and jitter, and uses a collective, abstracted data set derived from all network tests to detect outage events.

In measuring the impact of changing worker and consumer behaviour during COVID-19 lockdowns on internet performance, the research showed that not all services provider groups were the same.

ISPs were worst affected. Between January and July, cloud providers experienced around 400 outages globally compared with more than 4,500 in ISP networks. In terms of the total number of outages, 80 per cent occurred within ISP networks while fewer than 10 per cent happened among cloud providers. However, the research showed cloud provider outages may have been more serious in business terms.

"Despite fewer outages overall, the duration of outages compared to ISPs was either on-par or longer, indicating either maintenance activity or less agile recovery from network changes that unintentionally impacted availability," the report said.

Time was also a factor. Compared with ISP outages, a large portion of cloud outages, occur during peak business hours, potentially increasing their impact on users, it said.

In terms of geography, the Asia-Pacific region was hit hardest by ISP outages, with a 99 per cent increase compared with January. For North America, the figure was 65 per cent. ISP outages in EMEA did not increase so dramatically, but disruption grew in the following months and by June it was 45 per cent higher than in January. ®

SUBSCRIBE TO OUR WEEKLY TECH NEWSLETTER


Biting the hand that feeds IT © 1998–2020