Akamai Diversity

The Akamai Blog

Updated Insights into State of the Internet Metrics

In February 2015, we published a blog post entitled "State of the Internet Metrics: What Do They Mean?" which itself was an update to an earlier "Clarifying State of the Internet Report Metrics" blog post, published in March 2013.  The explanations in both posts are still relevant to the State of the Internet / Connectivity report series, but there are a few updates that are worth highlighting.

In the February 2015 post, we referenced "Broadband Connectivity", "High Broadband Connectivity", and "4K Readiness", which looked at the percentage of unique IPv4 addresses connecting to Akamai with average connection speeds above 4 Mbps, 10 Mbps, and 15 Mbps respectively.  However, during that same quarter, the United States FCC updated their definition of "broadband" from 4 Mbps to 25 Mbps, noting that the 4 Mbps standard, which had been established in 2010, was dated.  Not wanting to create confusion about differing definitions of the term "broadband" over time, we chose to instead begin referencing these metrics by way of their associated speeds.  As such, starting with the Second Quarter, 2015 State of the Internet Report, we began looking at 4, 10, 15, and 25 Mbps broadband adoption.  In adding the new 25 Mbps broadband adoption metric, we chose to look at it at a state level within the United States, since this is obviously the geography that the FCC's definition is most applicable to; we are also looking at 25 Mbps broadband adoption rankings globally, but are not including them within the regional sections. 

Over the last several years, we have analyzed and reviewed traffic from known mobile networks in its own section of the report.  As such, mobile network data has been removed from the data set used to calculate the metrics in the Global and United States sections, as well as subsequent regional "Geography" sections. However, a small number of network providers offer both fixed and mobile broadband service, and in some cases it may not be possible to accurately separate the two types of traffic within that network. This may result in the inclusion of some data in the 'fixed broadband' sections that is based on connections from mobile devices and/or mobile gateways. In the vast majority of cases, we do not expect this data to have a significant bearing on the results presented within the report, but in a few instances, the speeds presented may be substantively affected, and we will note within the body of the report those instances where we feel this may be the case.  

To that end, the mobile speed measurements included within the report - particularly average peak connection speeds - can be influenced by a number of factors, including the use and location of proxies within mobile networks. If a country's major mobile carriers make heavy use of such proxies, peak connection speeds recorded for that country are likely to be influenced by the speeds achieved between Akamai and the proxies (residing in data centers) rather than speeds achieved between Akamai and the mobile devices themselves. Some of the average peak connection speeds presented within the report are higher than one might expect given the current state of LTE and LTE-A deployment, so it is likely that these speeds are affected - to differing degrees - by proxies within those countries' mobile provider networks and may not be fully representative of speeds being seen by end users.

Finally, it is worth noting that some countries, such as Luxembourg, have chosen to roll out new high-speed broadband services using native IPv6 connectivity, and as such, the State of the Internet Report may under-report the connection speeds available to, and achieved by, broadband subscribers within these countries, as the speed calculations within the report are currently restricted to connections from IPv4 addresses only.  Moving beyond reporting IPv6 adoption statistics to including IPv6 connection speed insights remains on our to-do list.  However, since subscribers are generally assigned large blocks of IPv6 addresses, with different providers often assigning blocks of varying sizes, this makes generally applying the aggregation techniques we apply for IPv4 connection speed reporting more challenging for IPv6.  Addressing this challenge remains important to us, and we are still planning to integrate IPv6 connection speed reporting into the report in the future.

The latest insight into fixed and mobile connection speeds and broadband adoption can be found in the Second Quarter, 2016 State of the Internet Report, which will be published on September 29 on akamai.com.


Question: What causes the sharp spikes and declines in quarter measurements - (e.g. see South Africa 25Mbps Percentage)?

Hi David, can you explain what causes the sharp spikes and declines in quarter measurements.

For example, South Africa showed the percentage 25Mbps adoption as 0.8966% in Q4 15. In Q1 16, this jumped to 3.6%, but in Q2 16, this dropped to 2.6%.

The same observation applies to the average connection speed trend, I see peaks and dips.

From your view, what are the factors that cause such behaviour?

Kind regards



As we discuss in https://blogs.akamai.com/2015/02/state-of-the-internet-metrics-what-do-they-mean-1.html, the average connection speed metric can be influenced by a number of factors outside of Akamai's control. Within the report, we have also noted that changes to Akamai connectivity with one or more major network providers within a given country can also impact the reported speeds.

Given that the 25 Mbps broadband adoption metric (actually, all of the broadband adoption metrics) is tied to the average connection speed metric, changes in the latter can influence changes in the former.

In general, when talking with press and analysts about the data within the report, I suggest that the most important trend to look at is the long-term (year-over-year, or multi-year), as it gives the best picture of how things are progressing.