The State of Internet

VOLUME 6, NUMBER 1
1ST QUARTER, 2013 REPORT
TheState of
the Internet
INCLUDES INSIGHT ON MOBILE TRAFFIC AND CONNECTED DEVICES FROM ERICSSON

AQUA
Web
TERRA
Enterprise
KONA
Security
SOLA
Media
AURA
Network
Introducing the Complete Solution.
Hyperconnected. Cloud and Mobile Ready.
The hyperconnected world presents tremendous opportunities for businesses to lead.
Four key trends shape today’s marketplace: mobile, media, cloud, and security. Each one
represents its own set of challenges and opportunities for businesses. To help customers
accelerate growth in this landscape, Akamai continues to develop innovative solutions
that leverage the power of the Intelligent Platform.™
www.akamai.com/solutions
BUILT ON THE AKAMAI INTELLIGENT PLATFORMTM
SERVICE SUPPORT FROM INTERNET EXPERTS
Optimize and mobilize
to grow revenues
Gain business agility
by leveraging cloud
Web security so you can
innovate fearlessly
Engage your audience
globally and at scale
Monetize your network
and control costs
Control Center

Letter From the Editor
Over the last several years, Akamai has worked to provide additional insight, beyond the State of the Internet Report,
into what is happening on our Intelligent Platform through data visualization tools. The tools available at
www.akamai.com/stateoftheinternet endeavor to provide users with customized views of current and historical data
from the report, while others at www.akamai.com/ipv6 and www.akamai.com/html/technology/visualizing_akamai.html
provide a much broader perspective.
A year ago, we also launched Akamai IO, with the intention of providing actionable data on mobile browser usage
as seen across a sample of Akamai traffic. We have referenced updates to the data sets powering Akamai IO in last
quarter’s and this quarter’s reports, and we expect that an updated interface will also be available by the end of
August. With the new interface, available at www.akamai.com/io, users can filter and group the data across multiple
categories including device OS, browser, device type, network type, continent, and country, enabling unique and
customized views of the data.
In addition, we have also recently released a Web version of our “Akamai in 60 Seconds” visualization at
www.akamai.com/60seconds. In addition to serving some two trillion content requests a day, there’s even more
happening on the Akamai Intelligent Platform at any given time: video streaming, route optimization calculations,
DNS lookups, and content purges, just to name just a few. With so much taking place, “Akamai in 60 Seconds”
presents a snapshot of the broad range of activity occurring on the platform. The data points shown in the
visualization represent peak values for the given metrics as measured across the Akamai Intelligent Platform during
a 60-second time period, and are collected from several internal Akamai systems and databases, normalized to one
minute. Because the graphic is showing peak values over time, the numbers shown will change when new peaks are
reached, and it is likely that event-related activity will drive updates to some metrics.
We plan to continue to evolve and enhance both Akamai IO and Akamai in 60 Seconds throughout 2013, as well
as improve the State of the Internet Report with additional and better information in areas like attack traffic, mobile
connectivity, and IPv6, as well as updating design elements of the report. We’ve started with this quarter’s report, adding
an “Americas” section, highlighting connectivity metrics in countries across the North and South America continents.
As always, if you have questions, comments, or suggestions about the State of the Internet Report, connect with us
via e-mail at stateoftheinternet@akamai.com, or on Twitter at @akamai_soti.
– David Belson

SECTION 6: GEOGRAPHY — ASIA PACIFIC REGION 21
6.1 Asia Pacific Average Connection Speeds 21
6.2 Asia Pacific Average Peak Connection Speeds 22
6.3 Asia Pacific High Broadband Connectivity 22
6.4 Asia Pacific Broadband Connectivity 23
SECTION 7: GEOGRAPHY — EUROPE/MIDDLE EAST/AFRICA
(EMEA) 24
7.1 EMEA Average Connection Speeds 24
7.2 EMEA Average Peak Connection Speeds 24
7.3 EMEA High Broadband Connectivity 25
7.4 EMEA Broadband Connectivity 26
SECTION 8: MOBILE CONNECTIVITY 27
8.1 Connection Speeds on Mobile Networks 27
8.2 Mobile Browser Usage Data 29
8.3 Mobile Traffic Growth as Observed by Ericsson 30
SECTION 9: INTERNET EVENTS DISRUPTIONS 31
9.1 Hugo Chavez Dies 31
9.2 New Pope Chosen 31
9.3 Syria 31
9.4 SEACOM 33
9.5 SEA-ME-WE 4 Cable Cut 33
SECTION 10: APPENDIX 35
SECTION 11: ENDNOTES 36
EXECUTIVE SUMMARY 3
SECTION 1: SECURITY 4
1.1 Attack Traffic, Top Originating Countries 4
1.2. Attack Traffic, Top Ports 4
1.3 Observations on DDoS Attacks 5
1.4 “Account Checker” Attacks Target E-Commerce Sites 7
SECTION 2: INTERNET PENETRATION 8
2.1 Unique IPv4 Addresses 8
2.2 IPv4 Exhaustion 8
2.3 IPv6 Adoption 10
SECTION 3: GEOGRAPHY — GLOBAL 12
3.1 Global Average Connection Speeds 12
3.2 Global Average Peak Connection Speeds 12
3.3 Global High Broadband Connectivity 13
3.4 Global Broadband Connectivity 14
SECTION 4: GEOGRAPHY — UNITED STATES 15
4.1 United States Average Connection Speeds 15
4.2 United States Average Peak Connection Speeds 15
4.3 United States High Broadband Connectivity 16
4.4 United States Broadband Connectivity 17
SECTION 5: GEOGRAPHY — AMERICAS 18
5.1 Americas Average Connection Speeds 18
5.2 Americas Average Peak Connection Speeds 18
5.3 Americas High Broadband Connectivity 19
5.4 Americas Broadband Connectivity 20
Table of Contents
© 2013 Akamai Technologies, Inc. All Rights Reserved

5© 2013 Akamai Technologies, Inc. All Rights Reserved
Akamai’s globally-distributed Intelligent Platform allows us to gather massive amounts
of information on many metrics, including connection speeds, attack traffic, network
connectivity/availability/latency problems, and IPv6 growth/transition progress, as well
as traffic patterns across leading Web sites and digital media providers. Each quarter,
Akamai publishes the State of the Internet Report.
This quarter’s report includes data gathered from across the
Akamai Intelligent Platform in the first quarter of 2013 about
attack traffic, broadband adoption, and mobile connectivity,
as well as trends seen in this data over time. In addition, this
edition of the report includes insight into so-called “account
checker” attacks that targeted e-commerce sites, the states
of IPv4 exhaustion and IPv6 adoption, Internet “events” and
disruptions that occurred during the quarter, and observations
from Akamai partner Ericsson on data and voice traffic growth
on mobile networks.
Security
During the first quarter of 2013, Akamai observed attack traffic
originating from source IP addresses in 177 unique countries/
regions. Note that our methodology captures the source IP ad-
dress of an observed attack, and cannot determine attribution
of an attacker. China’s share of attack traffic fell to 34% during
the quarter, while Indonesia’s grew from near zero to over 20%.
Attack traffic from the United States fell from 10% to just over
8%. Attack traffic concentration grew significantly from the
fourth quarter of 2012, with the top 10 ports seeing 80% of
observed attacks. Significant growth was seen in attacks target-
ing Ports 80 (HTTP) and 443 (SSL), most of which came from
Indonesia. During the first quarter of 2013, Akamai’s customers
reported being targeted by 208 DDoS attacks, up 4% from
the prior quarter. Enterprise customers were most frequently
targeted, hit by 35% of the attacks. In addition, during the first
quarter Akamai observed attempted account takeover behavior
for a number of merchants resulting from reuse of credentials
obtained from other sites.
Internet and Broadband Adoption
Akamai observed a 3.1% increase in the number of unique
IPv4 addresses connecting to the Akamai platform, growing
to nearly 734 million, or approximately 34 million more than
were seen in the fourth quarter of 2012. Looking at connec-
tion speeds, the global average connection speed climbed
4.0% to 3.1 Mbps, and the global average peak connection
speed increased 9.2% to 18.4 Mbps. At a country level, South
Korea had the highest average connection speed at 14.2 Mbps,
while Hong Kong once again had the highest average peak
connection speed at 63.6 Mbps. Globally, high broadband (10
Mbps) adoption grew 10% quarter-over-quarter to 13%, and
South Korea remained the country with the highest level of
high broadband adoption, growing to 50%. Global broadband
(4 Mbps) adoption grew 5.8% to 46%, with Switzerland tak-
ing the top spot with 88% broadband adoption.
Mobile Connectivity
In the first quarter of 2013, average connection speeds on
surveyed mobile network providers ranged from a high of 8.6
Mbps down to a low of 0.4 Mbps. Average peak connection
speeds ranged from 45.6 Mbps down to 2.8 Mbps. Based on
traffic data collected by Ericsson, the volume of mobile data
traffic increased 19% between the fourth quarter of 2012
and the first quarter of 2013, while doubling year-over-year. In
contrast, mobile voice traffic grew only 4% during that same
year-over-year period.
Analysis of Akamai IO data collected across the first quarter
from a sample of requests to the Akamai Intelligent Platform
indicates that for users of devices on cellular networks, the
largest percentage of requests came from Android Webkit
(41-44%), ahead of Apple Mobile Safari (30-38%). However,
for users of mobile devices across all networks (not just cel-
lular), Apple Mobile Safari accounted for approximately 60%,
with Android Webkit responsible for 20-33% of requests.
(The ranges are related to updates made to the back-end data
source in the middle of the quarter.)
Executive Summary
5
Executive Summary
© 2013 Akamai Technologies, Inc. All Rights Reserved

6 © 2013 Akamai Technologies, Inc. All Rights Reserved
SECTION 1:
Security
Akamai maintains a distributed set of agents deployed across the Internet that monitor
attack traffic. Based on data collected by these agents, Akamai is able to identify the
top countries from which attack traffic originates, as well as the top ports targeted by
these attacks. Note that the originating country as identified by the source IP address
is not attribution — for example, a criminal in Russia may be launching attacks from
compromised systems in China. This section provides insight into port-level attack traffic,
as observed and measured by Akamai, during the first quarter of 2013. It also includes
insight into DDoS attacks that targeted Akamai customers during the first quarter of
2013, as well as insight into “account checker”-related attacks.
1.2 Attack Traffic, Top Ports
As shown in Figure 2, the concentration of attack traffic among
the top 10 targeted ports increased significantly during the first
quarter of 2013, driven primarily by significant increases in attack
volume targeting Ports 80 (WWW/HTTP) and 443 (SSL/HTTPS).
In fact, nearly 80% of the attacks targeting these ports were
observed to be originating in Indonesia, as referenced in Section
1.1. Despite these increases, Port 445 (Microsoft-DS) remained the
most targeted port, though the percentage of attacks targeting it
continued to decline, which is an encouraging trend. Of the top
10 targeted ports, Port 3389 (Microsoft Terminal Services) was
the only other one to see a decline quarter-over-quarter. Within
the list, Port 8080 (HTTP Alternate) was supplanted by Port 6882,
used unofficially by BitTorrent. All of the observed attacks target-
ing Port 6882 were observed to be originating in China. Data
from the Internet Storm Center1
shows a large spike in attacks
targeting this port late in the quarter; unfortunately, however,
there is no information provided on the source of the attacks.
1.1 Attack Traffic, Top Originating Countries
During the first quarter of 2013, Akamai observed attack
traffic originating from 177 unique countries/regions, consis-
tent with the count in the prior quarter. As shown in Figure
1, China remained the top source of observed attack traffic,
though its percentage declined by nearly a fifth from the
prior quarter. This decline is likely related to Indonesia making
a sudden appearance in the second place slot, after a 30x
increase quarter-over-quarter. The vast majority (94%) of the
attacks from Indonesia targeted Ports 80 (WWW/HTTP) and
443 (HTTPS/SSL), potentially indicating aggressive botnet activ-
ity. Hong Kong and India were the only two other countries/
regions among the top 10 that also saw quarterly increases in
observed attack traffic volume — the remaining countries/re-
gions saw nominal declines, in general. Attack traffic concen-
tration also increased in the first quarter, again owing to the
significant volume of attack traffic observed from Indonesia.
The makeup of the top 10 list remained largely consistent
with the previous quarter, with Italy and Hungary dropping off,
and Indonesia and Hong Kong joining.
In examining the regional distribution of observed attack
traffic in the first quarter, we find that nearly 68% originated
in the Asia Pacific/Oceania region, up from 56% in the fourth
quarter of 2012, likely due to the massive increase seen in
Indonesia. Europe accounted for just under 19%, while North
and South America originated just over 13% combined. Af-
rica’s contribution dropped as compared to prior quarters, as it
was responsible for a mere half a percent.
Figure 1: Attack Traffic,Top Originating Countries
(by source IP address, not attribution)
1 China 34% 41%
2 Indonesia 21% 0.7%
3 United States 8.3% 10%
4 Turkey 4.5% 4.7%
5 Russia 2.7% 4.3%
6 India 2.6% 2.3%
7 Taiwan 2.5% 3.7%
8 Brazil 2.2% 3.3%
9 Romania 2.0% 2.8%
10 Hong Kong 1.6% 1.2%
– Other 18% 25%
Q4 ’12 %Q1 ’13 % TrafficCountry

Port 445 remained the most targeted port in six of the top
10 countries and accounted for 70 times as much traffic as
the second most targeted port (135) in Romania — ratios in
the other countries ranged between 2 to 10 times as much.
In Turkey and Hong Kong, the largest number of attacks tar-
geted Port 23 (Telnet) — in previous quarters, this was the case
in Taiwan as well; however, in the first quarter, Port 445 was
targeted by approximately 5x as many attacks from Taiwan
as Port 23. (Interestingly, in the fourth quarter of 2012, Port
445 was not even among the top 10 ports targeted by attacks
originating in Taiwan.) The distribution of second-most tar-
geted ports was a bit broader in the first quarter, with Port 23
coming in second in Russia, Taiwan, and Brazil, and Port 1433
coming in second in India and Hong Kong. In the remaining
countries, the second spot was held by Port 3389 (China),
Port 443 (Indonesia), Port 80 (United States), Port 445 (Turkey),
and Port 135 (Romania).
1.3 Observations on DDoS Attacks
With the 4th Quarter, 2012 State of the Internet Report, Akamai
began analyzing the number of Distributed Denial of Service
(DDoS) attacks reported by our customers. Due to the nature
of the Akamai platform, it can be difficult to distinguish the
difference between a spike in the traffic a site might encounter
due to a specific event (such as being mentioned on a popular
live television program) and a malicious attack such as a DDoS,
without extensive review by an analyst. Therefore, Akamai relies
on customers to report DDoS attacks, which is what this analysis
is based upon. Additionally, attacks that target lower level
network layers, such as SYN floods, UDP floods and similar types
of volumetric attacks are not tracked by Akamai, as they are
automatically mitigated with minimal human involvement under
most circumstances. Higher level attacks that target the applica-
tion and logical layers, such as HTTP GET floods or attacks that
repeatedly download large files, are mitigated using Akamai’s
KONA Web Application Firewall solution, and require the inter-
vention of Akamai analysts to create, implement and disseminate
the rules to stop these attacks.
Across the full year 2012, 768 attacks were reported to Akamai,
and this shows little to no sign of abating in 2013. The fourth
quarter of 2012 saw 200 reported attacks, while 208 attacks
were reported in the first quarter of 2013, representing a slight
(4%) increase in the number of attacks reported. In the third
and fourth quarters of 2012, a significant number (72) of DDoS
attacks were attributed to the Izz ad-Dim al-Qassam Cyber Fight-
ers (aka QCF) and Operation Ababil. In the first quarter of 2013,
the tactics of these attacks changed, with the QCF no longer
announcing their targets prior to the attacks. Additionally, the
attacks ceased as of March 5, in theory to support a planned
operation known as “OpUSA” originating from members of the
group “Anonymous”. However it is unknown if this was truly
the case, or if the forces behind the QCF were merely pausing to
regroup for future attacks.
As illustrated in Figure 3, enterprise clients received a substan-
tially greater percentage of attacks in the first quarter of 2013,
accounting for 35% of all attacks (72 total), up 14% quarter
over quarter. The commerce and media verticals stayed relatively
Figure 2: Attack Traffic,Top Ports
445 Microsoft-DS 23% 29%
80 WWW (HTTP) 14% 2.8%
443 SSL (HTTPS) 11% 2.1%
23 Telnet 9.3% 7.2%
1433 Microsoft SQL Server 8.3% 5.3%
3389 Microsoft Terminal Services 5.4% 5.7%
3306 MySQL 2.7% 1.6%
22 SSH 2.6% 2.5%
135 Microsoft-RPC 2.2% 2.2%
6882 BitTorrent (unofficial) 1.5% –
Various Other 20% 40%
Q4 ’12 %Q1 ’13 % TrafficPort UsePort
Commerce
67
Public
Sector
9
Enterprise
72
High Tech
15
Media
Entertainment
45
Figure 3: Q1 2013 DDoS Attacks by Sector

This is directly related to the attacks performed by the QCF, as it
was in 2012. What is not apparent from the number of attacks is
the fact that a number of shorter, less impactful attacks were per-
formed in the first quarter, comprised of probes, rather than full-on
DDoS attacks. For victims (sites) that were affected by the attacks,
the aggressors would return at a later date to have a greater,
longer lasting impact on the target. These probes are often not ap-
parent until the full attack commences and are usually considered
to be part of the main attack for the purposes of this report, rather
than being recorded separately.
The media and entertainment sector continues to be a tempting
target, and was essentially unchanged (22%) as an overall percent-
age of DDoS attacks. Public sector customers suffered from statisti-
cally fewer attacks during the first quarter of the year, although
preliminary numbers for the second quarter of 2013 indicate that
this may be a temporary change to the state of affairs.
The number of DDoS attacks Akamai encounters shows every
indication of continuing to grow, with nearly 5% more attacks be-
ing reported in the first quarter of 2013 as compared to the fourth
quarter of 2012. It remains difficult to determine the nature of
the attackers because botnets are necessary to create the attacks
and the command and control (CC) infrastructures of these
botnets are designed to protect their owners. Another interest-
ing development is a return to the use of DNS reflection attacks.
This attack methodology allows attackers to make a relatively
SECTION 1:
Security (continued)
close to their 2012 percentages, at 32% vs. 34% for commerce
and 22% for media. At the same time, high tech and public
sector customers were targeted by substantially fewer attacks as
a percentage, at 7% and 4% of total attacks respectively. It is
interesting to note that the attacks in the first quarter were more
distributed (organizationally) than the attacks reported in 2012.
There were 154 unique organizations that reported DDoS attacks
in the first quarter, in contrast to 413 in all of 2012. This means
that nearly half (350) of the attacks in 2012 were against orga-
nizations that had already been attacked at least once, while this
number fell to 27% of attacks (54) in the first quarter of 2013.
The decline in the number of repeat targets may account for the
change in distribution of attacks.
As a percentage, first quarter attacks targeting the commerce sec-
tor remained relatively stable in comparison to the attacks reported
in 2012. While the distribution of the attacks remained nearly the
same, the actual targets were more varied, again following the
overall trend of spreading the targets of attacks across multiple
sites. As highlighted in Figure 4, retail organizations continue to
be tempting targets, primarily because they rely so heavily on the
Internet for sales and marketing and can be severely impacted if
their customers cannot reach their sites.
As shown in Figure 5, at the beginning of 2013, financial services
customers continued to bear the brunt of the attacks against the
enterprise vertical, suffering from 50% of all attacks in this vertical.
Hotel and Travel
16
Consumer
Goods
4
Retail
46
SaaS
1
Figure 4: Q1 2013 DDoS Attacks in the Commerce Sector
Financial Services
36
Miscellaneous
11
Business Services
15
Automotive
5
Energy
Utilities
5
Figure 5: Q1 2013 DDoS Attacks in the Enterprise Sector

The attackers then use their list of compromised accounts with
the installed tools to rapidly check the validity of the accounts.
Accounts that work are marked, and the attackers log in using
the credentials. Once logged in, the attackers can collect the
user’s personal data and credit card information to use for
further fraud. Attackers then modify the shipping address of the
victim and make purchases with their stored information. The
merchandise is sent to an address near the attacker and picked
up. In recent attacks, gift cards, both physical and electronic,
have been key items for purchase as they are readily available,
difficult to trace, and easy to transport.
The following are indications that an account checker has been
used against an e-commerce site:
• User complains that their account mailing address
has been altered
• Multiple other users altered in a similar time frame
• Many failed logins detected in a short period of time
from a small number of IP addresses
• Locked accounts
• Higher than normal rate of fraud activity
There are a number of defenses that can be used by merchants
to identify and stop these attacks. The tools used by the at-
tackers are not currently capable of bypassing CAPTCHA’s or
other validation requiring user intervention, which makes these
protections highly effective. Careful review of authentication
logs can be used to identify likely proxy servers used by the
attackers; a large number of sequential users from the same
IP address or group of IP addresses may be an indicator of this
type of an attack.
Akamai offers several tools that can used to combat this type
of attacker. The Akamai User Validation Module (UVM) can be
used to confirm that the login is coming from a browser and
therefore a live user. Merchants whose customer base is primar-
ily in known countries/regions can employ geo-blocking to limit
logins to users from specific geographies. In addition, organi-
zations that use Akamai’s Kona Site Defender can block these
attacks by using a combination of rate controls (how many con-
nections are allowed in a given time) and IP address block lists.
small investment in the traffic they send out while reaping a
huge reward in the amount of traffic sent to their target. By
sending a forged DNS request to an open, recursive DNS server,
attackers can easily multiply their attack traffic up to eight fold.
Due to poor Internet hygiene by many ISPs and the lack of
enforcement of BCP 38,2
forged DNS requests are allowed to
continue to the name servers, rather than being filtered by the
attacker’s ISP as they should be. For more information on this
topic, please refer to the DNS Reflection Defense blog post by
Akamai CSO Andy Ellis at https://blogs.akamai.com/2013/06/
dns-reflection-defense.html.
1.4 “Account Checker” Attacks Target E-Commerce Sites
In the first and second quarters of 2013, Akamai observed at-
tempted account takeover behavior for a number of merchants
resulting from reuse of credentials obtained from other sites.
Lists of username and password combinations are available in
carder forums or on pastebin, or acquired from compromised
merchants. Because users often use the same username and
password across multiple merchants and other non-commerce
sites, this allows attackers to use the compromised credentials
on a number of target merchants.
Attackers have been using automated tools — known as “ac-
count checkers” — to quickly determine valid user ID/password
combinations across a large number of e-commerce sites. Using
these tools, attackers can identify valid accounts rapidly and
gain access to the accounts, acquiring names, addresses and
credit card data from user profiles, and can also fraudulently
acquire merchandise.
Attackers usually begin by compromising hosted Web servers and
uploading a set of scripts to the compromised system in order
to use their resources in the attacks. Several known groups also
have semi-permanent or permanent domain names where they
host their tools. Akamai has also seen instances of attackers us-
ing cloud and virtual private server providers to host their attack
tools. These scripts make use of open proxies for the actual at-
tack; therefore it is not helpful to block the compromised systems.
Integral to the success of the attack is the use of Web proxies.
By routing traffic through open proxies, the attackers attempt
to bypass IP blocks. The tools Akamai has observed allow the at-
tackers to use a list of open proxies and cycle through them after
a fixed number of attempts. The attackers need to ensure their
list of proxies is both of sufficient size to disguise the attack and
that it contains valid proxies, or they risk exposing their attack.

2.1 Unique IPv4 Addresses
Through its globally-deployed Intelligent Platform, and by virtue
of the approximately two trillion requests for Web content
that it services on a daily basis, Akamai has unique visibility
into levels of Internet penetration around the world. In the
first quarter of 2013, over 733 million IP addresses, from 243
unique countries/regions, connected to the Akamai Intelligent
Platform — over 3% more than in the fourth quarter of 2012,
and 10% more than in the first quarter of 2012. Although we
see over 700 million unique IPv4 addresses, Akamai believes
that we see well over one billion Web users. In some cases,
multiple individuals may be represented by a single IPv4 address
(or a small number of IPv4 addresses), because they access the
Web through a firewall or proxy server, while in other cases, in-
dividual users can have multiple IPv4 addresses associated with
them due to their use of multiple connected devices. Unless
otherwise specified, the use of “IP address” within Section 2.1
refers to IPv4 addresses.
As shown in Figure 6, the global unique IPv4 address count
seen by Akamai grew by over 34 million quarter-over-quarter.
Quarterly growth was also seen among all of the top 10 coun-
tries, ranging from just 0.7% in Germany to 5.3% in China.
Looking at the full set of countries/regions around the world,
just shy of three-quarters of them saw a quarterly increase in
unique IP counts, with significant levels of growth experienced
in Angola, Sudan, and Kenya, which all grew in excess of 50%
quarter-over-quarter. (The driver of this growth isn’t immedi-
ately clear, though it could be related to improved Internet con-
nectivity becoming available in these countries from submarine
cables including EASSy3
and ACE.4
)
Looking at year-over-year changes, it is interesting to note
that China’s unique IP count has grown aggressively over time,
especially as compared to the United States. In January, it was
reported5
that the Internet population in China had grown
to 564 million users and was projected to grow to nearly 800
million users by 2015. It is not inconceivable that over the next
couple of years, this strong growth in China will enable its
unique IPv4 address count to eclipse that seen from the United
States. Other countries with strong year-over-year growth in
the first quarter included Italy, also with a 20% increase, and
Brazil, which grew at nearly double the rate of China and Italy,
at 38%. On a global basis, slightly less than three-quarters of
countries also had higher unique IP address counts year-over-
year, with several African and Middle Eastern nations having the
highest rates of growth.
2.2 IPv4 Exhaustion
As expected, the number of available IPv4 addresses continued to
decline heading into 2013, as Regional Internet Registries contin-
ued to allocate/assign blocks of IPv4 address space to organizations
within their respective territories.6
Leveraging data7
collected by
Geoff Huston,8
Chief Scientist at APNIC, the State of the Internet
Report can now provide a perspective on the size of the available
IPv4 address pool at each RIR, and how the pool sizes are shrinking
over time. In addition, the report will continue to use data pub-
lished by the RIRs to highlight IPv4 address assignment/allocation
activity by RIR over the course of each quarter.
Figure 7 illustrates the data provided by Mr. Huston, showing how
the size of the available IPv4 address pools at each of the RIRs
changed over the course of the first quarter of 2013. The smallest
rate of decline was seen at AFRINIC, which delegated just 0.7%
of its available space, and ended the quarter with just over 62.5
million available IPv4 addresses. As they are both delegating space
from their last “/8” address block, RIPE and APNIC unsurprisingly
had comparatively small drawdowns, handing out 3.5% and
1.9% of their available space, respectively. ARIN delegated 12.8%
of its available space, ending the quarter with just over 41.4 mil-
lion available IPv4 addresses, and LACNIC delegated a surprisingly
high 13.9% of available space, ending the first quarter with an
available pool of more than 62.5 million IPv4 addresses.
SECTION 2:
Internet Penetration
Figure 6: Unique IPv4 Addresses Seen by Akamai
– Global 733,799,401 3.1% 10%
1 United States 147,940,918 2.9% 1.0%
2 China 110,473,009 5.3% 20%
3 Japan 42,052,616 1.6% 3.8%
4 Germany 37,840,924 0.7% 5.0%
5 United Kingdom 28,524,028 3.3% 11%
6 France 26,992,978 1.9% 5.7%
7 Brazil 26,442,198 4.8% 38%
8 South Korea 21,412,948 2.5% 8.1%
9 Italy 20,232,010 1.6% 20%
10 Russia 18,240,078 3.1% 15%
YoY
Change
QoQ
Change
Q1 ’13 Unique
IPv4 Addresses
Country/Region

Figure 8 illustrates the IPv4 allocation/assignment activity across
each of the RIRs during the first quarter. As it shows, assignment
activity across APNIC, RIPE, and AFRINIC was fairly low, with the
exception of a small jump on February 12 in AFRINIC, representing
the assignment of a “/15” (131,072) IPv4 addresses to Telkom Mo-
bile,9
a South African mobile provider. Activity at ARIN saw several
significant jumps during the quarter, including over 650,000
addresses (a “/13” to Time Warner Cable Internet10
and a “/15” to
Akamai Technologies11
) handed out on January 24 and 25, over
900,000 IPv4 addresses delegated on February 19 (apparently to
Amazon.com12
), over one million on March 1 (to ATT Services,
Inc.13
), and another million on March 15 (also to Akamai Tech-
nologies14
). Activity at LACNIC also saw a number of significant
jumps during the quarter, including three blocks of over 262,000
addresses (a “/14” block to Brasil Telecom S/A,15
and two to Vivo
S.A,16,17
) on January 23 and 24, three similar delegations on Janu-
ary 30 (all apparently to Global Village Telecom18
), two delegations
of over 524,000 IPv4 addresses (a “/13” to the Administracion Na-
cional de Telecomunicaciones19
and one to Vivo S.A.20
) on February
19 and 20, and six “/14” blocks on March 14 (two to Tim Celular
S.A.,21
and four to NET Serviços de Comunicação S.A.22
) .
Figure 7: Available IPv4 Address Pool Size by RIR, Q1 2013
AFRINIC
APNIC
ARIN
LACNIC
RIPE
7
6
5
4
3
2
1
0
1-Jan
8-Jan
15-Jan
22-Jan
29-Jan
5-Feb
12-Feb
19-Feb
26-Feb
5-Mar
12-Mar
19-Mar
26-Mar
IPv4Addresses(Millions)
Figure 8: Total Number of IPv4 Addresses Assigned/Allocated by RIR, Q1 2013
8
7
6
5
4
3
2
1
0
IPv4Addresses(Millions)
1-Jan
8-Jan
15-Jan
22-Jan
29-Jan
5-Feb
12-Feb
19-Feb
26-Feb
5-Mar
12-Mar
19-Mar
26-Mar
AFRINIC
APNIC
ARIN
LACNIC
RIPE

SECTION 2:
Internet Penetration (continued)
2.3 IPv6 Adoption
As Akamai continues to roll out IPv6 support across its solu-
tion portfolio, we will endeavor to include data in the State of
the Internet Report on IPv6 adoption based on the analysis of
IPv6 requests to, and traffic delivered by, the Akamai Intel-
ligent Platform. However, until such time as we can include
comprehensive Akamai data on IPv6 adoption, we continue to
supplement with third-party data.
One such data source is network service provider Hurricane
Electric, which claims that it is “considered the largest IPv6
backbone in the world as measured by number of networks
connected.”23
A white paper24
available from Hurricane Elec-
tric highlights more than a decade of experience in working
with IPv6, and they also publish the “Global IPv6 Deployment
Progress Report”, available at http://bgp.he.net/ipv6-progress-
report.cgi.
As is evident in Figure 9, the rate of growth in the first quarter
of 2013 was significantly lower than that seen in the previous
four years — in fact, the 5.0% increase was half as much as in
2012, and a quarter of that seen in 2011. Just 333 Autono-
mous Systems (ASes) were added to the IPv6 routing table in
the first quarter of 2013, as compared to 489 in 2012 and
589 in 2011. However, 2013’s figure was larger than the num-
ber added in both 2010 and 2009. While this metric provides
some perspective around global IPv6 adoption, it is also impor-
tant to recognize that not all ASes are equivalent — ASes as-
sociated with large numbers of users/subscribers are ultimately
more meaningful for measuring the success of IPv6 adoption
than ASes not directly associated with end user connectivity/
traffic. Data collected by Akamai and others indicates that
ASes used for end user traffic are starting to see meaningful
levels of IPv6 adoption.25
Since mid-2012, we have been tracking IPv6 traffic to the
Akamai Intelligent Platform. The graph in Figure 10 is based
on data taken from http://www.akamai.com/IPv6, which pro-
vides both rolling 24-hour and historical views of IPv6 request
volume seen by Akamai (in hits/second), and shows peak
traffic volume on a daily basis across the first quarter of 2013.
IPv6 traffic continues to exhibit a somewhat cyclical weekly
pattern, as was noted last quarter as well. Absolute volume
grew from almost 92,000 hits/second at the start of 2013 to
just over 115,000 hits/second at the end of the first quarter.
There is a very interesting shift in the peaks and troughs of the
weekly cycle that is observable starting in early March as well,
with differences of 20,000 hits/second or more evident during
several weeks — it is not clear exactly what drove this change.
Figure 9: Total Number of Autonomous Systems in the IPv6 Routing Table
2013
2012
2011
2010
2009
8
7
6
5
4
3
2
1
0
AutonomousSystems(Thousands)
1-Jan
8-Jan
15-Jan
22-Jan
29-Jan
5-Feb
12-Feb
19-Feb
26-Feb
5-Mar
12-Mar
19-Mar
26-Mar

DID YOU KNOW?
• In the IPv6 network operator statistics update published by the Internet Society in March 2013, tens
of additional networks qualified for listing. The Internet Society noted that Deutsche Telekom, Tele-
fonica del Peru, and VOO from Belgium are all new entrants that were already in the top 12 opera-
tors listed, meaning they were already delivering services for significant numbers of IPv6 users.
[Source: http://www.worldipv6launch.org/more-operators-more-ipv6-2013-off-to-a-flying-start/]
• You can follow the Regional Internet Registries on Twitter to find out more about local IPv4
exhaustion and IPv6 adoption:
• AFRINIC: @AFRINIC • LACNIC: @LACNIC
• APNIC:@APNIC • RIPE: @RIPE_NCC
• ARIN: @TeamARIN
• RIPE, the European RIR, provides insight into the percentage of IPv6-enabled networks at a
country or group of countries level at http://v6asns.ripe.net/v/6
140
120
100
80
60
40
20
0
Hits/Second(Thousands)
1-Jan
8-Jan
15-Jan
22-Jan
29-Jan
5-Feb
12-Feb
19-Feb
26-Feb
5-Mar
12-Mar
19-Mar
26-Mar
Figure 10: IPv6 Traffic to the Akamai Intelligent Platform, Q1 2013

a total of 123 qualifying countries/regions saw a year-over-year
increase in average connection speeds, ranging from just 1.4% in
Oman (to 1.2 Mbps) to a surprisingly high 122% in Iraq (to 2.0
Mbps). Year-over-year declines in average connection speeds were
seen in just 15 countries/regions, ranging from a loss of just 0.6%
in Panama (to 2.8 Mbps) to a drop of 19% in Egypt (to 1.1 Mbps).
In the first quarter, 14 qualifying countries/regions had average
connection speeds of 1 Mbps or less, down from 18 in the prior
quarter. This continued decline is an encouraging trend, likely
pointing to improved broadband connectivity across even the
slowest geographies. Libya once again held the spot as the coun-
try with the lowest average connection speed, clocking in at 0.6
Mbps, growing just less than one percent quarter-over-quarter, but
19% year-over-year.
3.2 Global Average Peak Connection Speeds
The average peak connection speed metric represents an
average of the maximum measured connection speeds across
all of the unique IP addresses seen by Akamai from a particu-
lar geography, and is more representative of Internet connec-
tion capacity. The average is used to mitigate the impact of
unrepresentative maximum measured connection speeds.
The data presented within this section was collected during the first quarter of 2013
through Akamai’s globally-deployed Intelligent Platform and includes all countries that
had more than 25,000 unique IP addresses make requests for content to Akamai dur-
ing the quarter. For purposes of classification within this report, the “high broadband”
data included below is for connections at greater than 10 Mbps, and “broadband” is
for connections of 4 Mbps or greater.
SECTION 3:
Geography –Global
– Global 3.1 4.0% 17%
1 South Korea 14.2 1.5% -10%
2 Japan 11.7 3.9% 6.8%
3 Hong Kong 10.9 9.0% 16%
4 Switzerland 10.1 6.1% 24%
5 Netherlands 9.9 10% 12%
6 Latvia 9.8 4.5% 12%
7 Czech Republic 9.6 9.1% 34%
8 Sweden 8.9 10.8% 32%
9 United States 8.6 7.4% 27%
10 Denmark 8.2 13% 17%
YoY
Change
QoQ
Change
Q1 ’13
Avg. Mbps
Country/Region
Figure 11: Average Measured Connection Speed by Country/Region
In addition to providing insight into high broadband and broad-
band adoption levels, the report also includes data on average
and average peak connection speeds — the latter provides insight
into the peak speeds that users can likely expect from their Inter-
net connections.
Finally, traffic from known mobile networks is analyzed and
reviewed in a separate section of the report; mobile network
data has been removed from the data set used to calculate the
metrics in the present section, as well as subsequent regional
“Geography” sections.
3.1 Global Average Connection Speeds
The global average connection speed continued to grow in the
first quarter of 2013, increasing 4.0% to 3.1 Mbps. As shown
in Figure 11, increases were seen among all of the top 10
countries/regions, with particularly strong increases seen in the
Netherlands, Sweden, and Denmark — all three European coun-
tries grew 10% or more quarter-over-quarter. Overall, among
the top 10, growth ranged from just 1.5% in South Korea to
13% in Denmark. While South Korea maintained its position as
the country with the highest average connection speed, both
Hong Kong and Switzerland posted large enough quarterly gains
to push them over the “high broadband” 10 Mbps threshold
as well. Globally, a total of 117 countries/regions that qualified
for inclusion saw average connection speeds increase from the
fourth quarter of 2012, ranging from 0.7% in Kuwait (to 2.1
Mbps) to a massive 75% in Guatemala (to 3.2 Mbps). Quarterly
losses ranged from 0.1% in Tunisia (to 1.5 Mbps) to 13% in
Bahrain (to 1.9 Mbps).
Long-term trends are generally rather positive as well. At a global
level, the average connection speed is up 17% year-over-year, and
eight of the top 10 countries/regions also grew by double-digit
percentages, led by the Czech Republic and Sweden, which had
yearly growth of 34% and 32% respectively. Only Japan had a
growth rate below 10%, and South Korea was the only country
of the set that saw a year-over-year decline. Around the world,

As shown in Figure 12, the global average peak connection speed
saw very strong growth during the first quarter of 2013, increas-
ing 9.2% to 18.4 Mbps. A quarterly increase of 9.0% pushed first
place Hong Kong up past 60 Mbps for the first time, as it reached
an average peak connection speed of 63.6 Mbps. Japan’s solid
13% quarter-over-quarter change pushed it to 50 Mbps, making
it only the second country to reach that threshold. Five additional
countries among the top 10 joined Japan with double-digit per-
centage quarterly growth, with the Netherlands seeing the biggest
increase, at 15%. On a global basis, 130 qualifying countries/
regions saw higher average peak connection speeds quarter-over
quarter, with increases ranging from 0.9% in Nicaragua (to 13.3
Mbps) to 55% in Ghana (to 21.3 Mbps). Only six qualifying coun-
tries/regions saw lower average peak connection speeds quarter-
over-quarter, with losses ranging from just 0.4% in Indonesia (to
12.8 Mbps) to 7.9% in Sudan (to 6.1 Mbps).
Looking at year-over-year changes, the long term trends are very
strong as well. The global average peak connection speed was up
by 36%, and it was joined by similarly strong growth rates among
nine of the top 10 countries/regions, led by Singapore and Switzer-
land. Average peak connection speeds in both countries grew by
more than 40% — Singapore was up 43% and Switzerland 41%
year-over-year. Around the rest of the world, all but six of the quali-
fying countries/regions saw yearly growth in average peak connec-
tion speeds, ranging from 3.7% in Chile (to 20.4 Mbps) to 152%
in Indonesia. Mozambique also posted a year-over-year increase of
over 100%, gaining 107% (to 9.6 Mbps). Of the countries/regions
that saw yearly declines, losses ranged from 3.8% in Guatemala (to
15.6 Mbps) to 28% in Sudan. The lowest average peak connection
speed in the first quarter was recorded in Iran, at 3.2 Mbps — up
17% quarter-over-quarter, but down 7.1% year-over-year.
3.3 Global High Broadband Connectivity
In line with the strong growth seen in the connection speed
metrics, Figure 13 shows that high broadband adoption rates
followed suit in the first quarter, with the global high broadband
adoption rate growing to 13%, thanks to a 10% quarterly
increase. Among the top 10 countries/regions, seven also grew
10% or more quarter-over-quarter, with an impressive 25%
quarterly increase in the Netherlands topping the list. Quarterly
growth in the remaining three countries on the list was nominal,
with South Korea once again seeing half of its connections to
Akamai at speeds above 10 Mbps. In a similar vein, all of the top
ten countries are now seeing at least one in five connections to
Akamai at speeds above 10 Mbps — an encouraging sign. Of the
50 countries/regions around the world that qualified for inclusion,
only Thailand and China saw high broadband adoption rates
decline quarter-over-quarter, losing 0.5% (to 2.7% adoption)
and 21% (to 0.2% adoption) respectively. Quarterly growth in
the other countries/regions ranged from a slight 0.8% increase in
Australia (to 4.8% adoption) to a surprisingly large 102% jump
in India (to 0.3% adoption). With India’s massive increase, China
returned to its position as the country with the lowest level of
high broadband adoption among qualifying countries.
Looking at year-over-year changes, the global high broadband
adoption rate saw strong growth, up 28%. Among the top 10
countries/regions, nine saw extremely strong improvement as
compared to the first quarter of 2012, with growth rates ranging
from 17% in Japan, to a near-doubling in the Czech Republic,
which added 96%. Yearly changes among the qualifying coun-
tries/regions around the world ranged from growth of just 2.1%
in Belgium (to 19% adoption) to increases of over 200% in both
Moldova (up 227% to 16% adoption) and the United Kingdom
– Global 18.4 9.2% 36%
1 Hong Kong 63.6 9.0% 29%
2 Japan 50.0 13% 26%
3 Romania 47.9 8.9% 23%
4 South Korea 44.8 1.5% -6.3%
5 Latvia 44.2 10% 32%
6 Singapore 41.1 8.8% 43%
7 Switzerland 40.3 12% 41%
8 Bulgaria 38.2 14% 39%
10 Belgium 38.0 14% 30%
YoY
Change
QoQ
Change
Q1 ’13
Peak Mbps
Country/Region
Figure 12: Average Peak Connection Speed by Country/Region
– Global 13% 10% 28%
1 South Korea 50% 1.5% -6.2%
2 Japan 43% 5.7% 17%
3 Hong Kong 34% 10% 20%
4 Latvia 31% 6.3% 19%
5 Switzerland 30% 11% 58%
6 Netherlands 29% 25% 21%
7 Sweden 25% 18% 61%
8 United States 25% 14% 69%
9 Czech Republic 23% 10% 96%
10 Finland 21% 12% 30%
YoY
Change
QoQ
Change
% Above
10 Mbps
Country/Region
Figure 13: High Broadband (10 Mbps) Connectivity

SECTION 3:
Geography –Global (continued)
(up 208% to 20% adoption). Nine additional countries grew by
more than 100% year-over-year, while Greece fell just short, with
a year-over-year change of 99%. In addition to South Korea, yearly
declines were seen in Portugal (down 11% to 6.0% adoption),
and Chile (down 12% to 0.9% adoption).
3.4 Global Broadband Connectivity
In the first quarter of 2013, the global broadband adoption rate
saw nominal growth from the prior quarter, reaching 46% on
an increase of 5.8%. As shown in Figure 14, similar nominal
growth was seen across the top 10 countries/regions, though
Austria and Denmark stood out with quarter-over-quarter
changes above 10%, increasing 16% and 11% respectively. The
smallest change among the group was seen in the Netherlands,
which grew just 1.2%. Across the countries/regions in the top
10, all of them had at least three of every four requests to Aka-
mai at speeds above 4 Mbps. Globally, a total of 72 countries/re-
gions that qualified for inclusion saw higher levels of broadband
adoption quarter-over-quarter, with growth ranging from 1.2%
in the Netherlands to an unusual 108% increase in Réunion
(to 27% adoption). Thirty-seven countries/regions around the
world saw at least half of their connections to Akamai occurring
at speeds above 4 Mbps. In contrast, Venezuela had the lowest
level of broadband adoption in the first quarter, at just 1.2%.
Broadband adoption grew year-over-year at a global level,
increasing 12%, and was also up across all of the top 10
countries/regions, though the changes were not as large as
those seen for the high broadband metric. Among the group,
yearly increases ranged from just 1.2% in the Netherlands to
28% in Austria, while Switzerland, the Czech Republic, Canada,
and Denmark all joined Austria in growing more than 10%
year-over-year. Looking across the whole world, a total of 69
countries/regions saw higher broadband adoption levels year-
over-year. Growth rates ranged from the 1.2% increase in the
Netherlands to a massive 716% increase in Morocco (to 6.6%
adoption). In addition to Morocco, yearly growth in excess of
200% was also seen in Indonesia (up 414% to 3.7% adoption),
Réunion (up 330% to 27% adoption), Georgia (up 237% to
20% adoption), and Ecuador (up 218% to 13% adoption). Lev-
els of broadband adoption more than doubled in an additional
eight countries as compared to the first quarter of 2012. Only
seven qualifying countries saw lower levels of broadband adop-
tion year-over-year, with Vietnam’s 48% loss (to 1.9% adop-
tion) the largest seen.
Unsurprisingly, broadband access continues to be seen as a
global priority. Dr. Hamadoun Touré, the Secretary-General of
the International Telecommunication Union (ITU), told the 7th
Meeting of the Broadband Commission in Mexico City that he’d
like to “dream big” and set a new goal to ensure that broad-
band Internet speeds of 20Mbps for $20 a month are accessible
to everyone in the world by 2020.26
However, it is interesting to
note that this goal, while admirable, falls short of more localized
goals, such as the EU Digital Agenda goal of 30 Mbps connectiv-
ity for EU households by 2020 and the United Kingdom’s goal of
25+ Mbps connectivity for 90% of citizens by 2015.Figure 14: Broadband (4 Mbps) Connectivity
– Global 46% 5.8% 12%
1 Switzerland 88% 2.5% 14%
2 South Korea 87% 1.5% 1.5%
3 Netherlands 84% 1.2% 1.2%
4 Czech Republic 81% 8.7% 19%
5 Japan 79% 3.8% 9.9%
6 Hong Kong 78% 3.3% 8.4%
7 Austria 78% 16% 28%
8 Canada 77% 3.9% 13%
9 Denmark 77% 11% 14%
10 Belgium 76% 4.5% 5.1%
YoY
Change
QoQ
Change
% Above
4 Mbps
Country/Region
DID YOU KNOW?
A report published by the International Telecommunication Union (ITU) comparing broadband prices
in countries around the world highlighted that there are 17 countries that have prices equal to 100%
or more of average monthly income, while 25 countries have a price equal to 1% or less of average
monthly income.
[Source: http://royal.pingdom.com/2013/03/12/broadband-prices/]

4.1 United States Average Connection Speeds
The first quarter of 2013 brought extremely strong quarter-
over-quarter increases in average connection speed for the top
10 states, as illustrated in Figure 15. Vermont remained the
fastest state, at 12.7 Mbps, with a quarterly increase of 18%.
However, it wasn’t the largest increase among the group, as
Virginia added 22%, reaching an average connection speed
of 10.7 Mbps. Impressively, all of the states among the top 10
had average connection speeds above the “high broadband”
10 Mbps threshold in the first quarter, and all of them saw
quarterly growth of more than 10% as well. Across the whole
country, average connection speeds were up in every state, with
increases ranging from 3.2% in Minnesota (to 7.9 Mbps) to
an unexpectedly high 34% in North Dakota (to 9.6 Mbps). In
addition to North Dakota, Ohio, South Dakota, and Virginia all
saw quarterly increases above 20%. Unfortunately, Arkansas
continued to maintain its position as the state with the lowest
average connection speed, despite growing 16% to 4.4 Mbps.
On a year-over-year basis, growth was extremely strong among
the top 10 states, which saw increases ranging from 17% in
Delaware to 40% in Vermont — Delaware was the only state on
the list that saw growth lower than 20%. Solid year-over-year
growth was also seen across the rest of the country, with North
Dakota and Ohio surpassing Vermont’s aggressive growth,
increasing 43% (to 9.6 Mbps) and 42% (to 8.2 Mbps) respec-
tively. A total of 13 states saw growth of 30% or more, while
another 26 increased 20% or more year-over-year. Nevada
was the only state that increased less than 10% year-over-year,
growing 8.1% (to 7.4 Mbps).
4.2 United States Average Peak Connection Speeds
The first quarter of 2013 marks the first time that all of the
top 10 states had average peak connection speeds above 40
Mbps, as Figure 16 shows. The District of Columbia led the
pack, with an average peak connection speed of 47.2 Mbps,
up 9.4%. Vermont was just 86 kbps behind, with an average
peak connection speed of 47.1 Mbps, up 14% quarter-over-
quarter. Utah had the lowest average peak connection speed
among the top 10 states, at 41.9 Mbps, while Delaware had
the lowest quarterly increase, at 5.8%. Across the whole coun-
try, 33 states grew by 10% or more quarter-over-quarter, while
the lowest quarterly increase was seen in Nevada, at 4.2% (to
31.3 Mbps). A total of 14 states had average peak connection
speeds of 40 Mbps or more in the first quarter, while another
26 came in above 30 Mbps. Ten more had average peak con-
nection speeds above 20 Mbps, and the only state that had an
average peak connection speed below 20 Mbps was Arkansas,
at 19.8 Mbps (up 13% quarter-over-quarter.)
SECTION 4:
Geography –United States
The metrics presented here for the United States are based on a subset of data used
for Section 3 and are subject to the same thresholds and filters discussed within that
section. The subset used for this section includes connections identified as coming
from networks in the United States, based on classification by Akamai’s EdgeScape
geolocation tool.
Figure 15: Average Connection Speed by State
1 Vermont 12.7 18% 40%
2 New Hampshire 12.0 18% 28%
3 Delaware 11.9 12% 17%
4 District Of Columbia 11.3 11% 28%
5 Utah 11.0 16% 29%
6 Massachusetts 10.7 15% 31%
7 Virginia 10.7 22% 36%
8 Maryland 10.6 13% 31%
9 New Jersey 10.5 15% 34%
10 Connecticut 10.4 13% 24%
YoY
Change
QoQ
Change
Q1 ’13
Avg. Mbps
State
Figure 16: Average Peak Connection Speed by State
1 District Of Columbia 47.2 9.4% 37%
2 Vermont 47.1 14% 33%
3 New Jersey 45.7 19% 42%
4 New Hampshire 44.4 18% 32%
5 Massachusetts 43.8 14% 34%
6 Maryland 43.8 16% 37%
7 New York 43.1 12% 33%
8 Virginia 43.1 16% 27%
9 Delaware 42.8 5.8% -1.5%
10 Utah 41.9 14% 34%
YoY
Change
QoQ
Change
Q1 ’13
Peak Mbps
State

Year-over-year changes among the top 10 states were generally
very strong, with eight of the 10 above 30%. Delaware was
the only state among the group to see a long-term decline in its
average peak connection speed, dropping 1.5% from the first
quarter of 2012 — it’s not clear why, after increasing for several
quarters, the long-term trend in Delaware has turned negative.
Among all of the states in the union, year-over-year changes
were positive everywhere but Delaware. Kentucky saw the high-
est level of year-over-year growth, and was the only state with a
yearly increase above 50%, growing 52% (to 31.9 Mbps). Over-
all, a total of 17 states grew by 30% or more, while another 19
grew by 20% or more. The smallest yearly increase was seen in
Nevada, which added 10% (to 31.3 Mbps).
4.3 United States High Broadband Connectivity
Quarterly changes in high broadband adoption among the top 10
states followed the connection speed metrics in showing strong
growth in the first quarter, as evidenced in Figure 17. Surprisingly
high rates of quarterly growth were seen across these states, rang-
ing from 25% in New York to 42% in New Hampshire. In New
Hampshire, nearly half of connections to Akamai were at speeds
above 10 Mbps; in New York, just over one in three connected at
10 Mbps or more. All but three states across the whole country
had high broadband adoption rates above 10% — those three are
Hawaii (9.3% adoption, up 21%), Idaho (7.5% adoption, up 10%),
and Arkansas, which had the lowest level of high broadband adop-
tion in the country (4.4%, up 32%).
Year-over-year changes across the top 10 states were also extremely
strong, ranging from a low of 27% in Delaware, to a doubling
(100%) in New Jersey. Impressively, Delaware was the only state
among the top 10 that saw a yearly growth rate below 50%. Year-
over-year changes were also particularly high when looking at the
whole country. Three states saw growth over 100% — Missouri (up
120% to 16% adoption), Michigan (up 117% to 26% adoption),
and New Jersey. Kentucky fell just short, growing 99% year-over
year (to 12% adoption). In addition to these four states, an addi-
tional 37 saw high broadband adoption rates grow more than 50%
from the same period a year prior. Maine had the lowest yearly
increase in the first quarter, growing 21% (to 14% adoption).
High broadband adoption rates may continue to see aggressive
growth over the next couple of years if broadband providers and lo-
cal/state governments come together in support of the “Gigabit City
Challenge” issued by outgoing FCC Chairman Julius Genachowski
in January 2013.27
At the U.S. Conference of Mayors in January, Mr.
Genachowski called for “at least one gigabit community in all 50
states by 2015.” To help communities achieve this goal, he also an-
nounced plans “to create a new online clearinghouse of best prac-
tices to collect and disseminate information about how to lower the
SECTION 4:
Geography –United States (continued)
1 New Hampshire 48% 42% 65%
2 New Jersey 45% 36% 100%
3 Vermont 43% 36% 61%
4 Massachusetts 41% 29% 74%
5 Delaware 41% 30% 27%
6 District Of Columbia 41% 23% 52%
7 Maryland 40% 32% 74%
8 Rhode Island 40% 25% 64%
9 Connecticut 35% 32% 65%
10 New York 35% 25% 80%
YoY
Change
QoQ
Change
% Above
10 Mbps
State
Figure 17: High Broadband (10 Mbps) Connectivity, U.S. States
DID YOU KNOW?
Gigabit broadband connectivity is already available in a number of U.S. cities, including:
• Kansas City, KS / MO • Springfield, VT
• Chattanooga, TN • Omaha, NE
• Lafayette, LA • Tullahoma, TN
• East Lansing, MI • Minneapolis, MN
• Bristol, VA / TN • Cedar Falls, Iowa
• Morristown, TN • Seattle, WA
• Burlington, VT
[Source: http://highspeedgeek.com/america-gigabit-internet/]

costs and increase the speed of broadband deployment nationwide,
including to create gigabit communities.” Some communities are
already taking steps to become “gigabit communities” by working
with initiatives such as Gig. U. In February, Gig. U announced that
it had helped the North Carolina Next Generation Network, which
includes four universities (North Carolina State, University of North
Carolina-Chapel Hill, Duke and Wake Forest) and six communities,
put together an RFP for a project that they are hoping will entice
both existing and new broadband providers to bid on building and
running a network offering 1 Gbps, with a goal to start offering
service within 18 months.28
4.4 United States Broadband Connectivity
Although still all positive, the quarterly changes in broadband
adoption levels among the top 10 states were much lower than
those seen for high broadband connectivity, or for the con-
nection speed metrics. Interestingly, as shown in Figure 18, the
changes were fairly tightly clustered, ranging from 3.0% in
Delaware to 5.4% in Florida. Broadband adoption rates among
the top 10 states are also very strong, with at least three of every
four requests (and more in most cases) to Akamai from these
states made at speeds above 4 Mbps. The spread of quarter-over-
quarter changes across the whole country is broader, but tops
out at 24% in Arkansas. However, at 33% adoption, Arkansas
also ranks as the state with the lowest level of broadband adop-
tion in the first quarter. In addition to Arkansas, nine additional
states had quarterly growth rates above 10%, while the balance
all grew less than that.
1 New Hampshire 90% 4.0% 6.8%
2 Delaware 90% 3.0% -2.7%
3 Rhode Island 87% 5.0% 7.3%
4 Vermont 86% 4.9% 6.5%
5 New Jersey 84% 4.9% 12%
6 Maryland 82% 4.9% 15%
7 New York 82% 5.2% 15%
8 Connecticut 81% 4.9% 14%
9 Massachusetts 77% 3.6% 11%
10 Florida 76% 5.4% 14%
YoY
Change
QoQ
Change
% Above
4 Mbps
State
Figure 18: Broadband (4 Mbps) Connectivity, U.S. States
Year-over-year changes among the top 10 states, while generally
positive, were also less aggressive than those seen in the other
metrics. New Hampshire, Rhode Island, and Vermont grew less
than 10% on a yearly basis, while increases in the other states
on the list topped out at 15%, in Maryland and New York.
Across the whole country, Kentucky had the highest rate of
yearly growth, at 66% (to 64% adoption) – well ahead of Arkan-
sas, which had the next highest yearly increase, at 36%. In total,
14 states improved broadband adoption levels by 20% or more
year-over-year, while an additional 27 saw improvements of 10%
or more. Aside from Delaware’s 2.7% yearly decrease, Vermont
had the smallest rate of growth, at 6.5%.
DID YOU KNOW?
• The 2013 Measuring Broadband America—February Report from the United States Federal Com-
munication Commission made three primary observations regarding the current state of residential
broadband service in the United States:
• Many ISPs continue to closely meet or exceed the speeds they advertise.
• Consumers are continuing to migrate to faster speed tiers.
• Satellite broadband has made significant improvements in service quality.
• The 2013 Measuring Broadband America—February Report also found that on average, during peak
periods DSL-based services delivered download speeds that were 85 percent of advertised speeds,
cable-based services delivered 99 percent of advertised speeds, fiber-to-the-home services delivered
115 percent of advertised speeds, and satellite delivered 137 percent of advertised speeds.
[Source: http://www.fcc.gov/measuring-broadband-america/2013/February]

SECTION 5:
Geography –Americas
5.1 Americas Average Connection Speeds
In the first quarter of 2013, the United States and Canada had
the highest average connection speeds among countries in the
Americas, as shown in Figure 19. Measured connection speeds
of 8.6 Mbps and 7.8 Mbps respectively were more than double
that of neighboring Mexico. Within Latin and South America,
average connection speeds ranged from 3.3 Mbps in Mexico to
just 0.9 Mbps in Bolivia. Quarter-over-quarter changes among
the surveyed countries were generally positive, though fairly
nominal. Only Bolivia saw its average connection speed drop
quarter-over-quarter (by 0.7%), while the other countries saw
increases ranging from 1.5% in Colombia to 9.2% in Mexico.
On a year-over-year basis, changes were, surprisingly, more
mixed, though the majority of surveyed countries saw yearly
growth in average connection speeds. Among the surveyed
countries that saw increases, the growth was fairly strong,
ranging from 7.4% in Brazil to 51% in Bolivia; Brazil was the
only country with growth below 10%. The observed quarterly
declines were fairly nominal — Paraguay lost 3.6%, Argentina
was down 6.6%, and Chile dropped 11% year-over-year.
Interestingly, Bolivia and Venezuela, the two countries with the
lowest average connection speed in the first quarter, recently
renationalized their incumbent telecommunications providers.29
Bolivia’s rank as the country with the lowest average connection
speed may be related to the high cost of Internet connectivity
there — a 1 Mbps connection costs 55% of GDP per capita. In
contrast, the same connection costs 1.3% of GDP per capita in
Uruguay and a fraction of a percent in European nations such
as Spain and France.30
5.2 Americas Average Peak Connection Speeds
Similar to the gap seen in the average connection speed metric,
the United States and Canada were far ahead of the other sur-
veyed Americas countries in the average peak connection speed
metric as well, as demonstrated in Figure 20. The United States
topped the list at 36.6 Mbps, up a solid 11% quarter-over-quar-
ter, while Canada posted a 34.2 Mbps average peak connection
speed, growing 13% from the fourth quarter of 2012. Chile
was the only other country in the Americas region that came in
above 20 Mbps, though just barely, while nine more countries
had average peak connection speeds above 10 Mbps. The low-
est average peak connection speed among surveyed countries
was seen in Venezuela, at 8.2 Mbps, up 8.8% quarter-over-
quarter. Similar to Venezuela’s quarterly growth, changes seen
in other surveyed countries were similarly as moderate, ranging
from a low of 2.9% in Chile to a high of 17% in Mexico.
YoY
Change
QoQ
Change
Q1 ’13
Avg. Mbps
Country/Region
9 United States 8.6 7.4% 27%
13 Canada 7.8 7.8% 21%
57 Mexico 3.3 9.2% 19%
63 Chile 3.0 2.7% -11%
64 Colombia 2.8 1.5% 5.8%
72 Ecuador 2.3 2.8% 34%
73 Brazil 2.3 4.4% 7.4%
78 Costa Rica 2.1 5.7% 19%
82 Argentina 2.1 4.7% -6.6%
85 Peru 2.0 5.7% 24%
99 Uruguay 1.7 4.0% 29%
117 Paraguay 1.2 6.3% -3.6%
123 Venezuela 1.1 5.1% 19%
128 Bolivia 0.9 -0.7% 51%
Global
Rank
Figure 19: Average Connection Speed by Americas Country
The metrics presented here for the Americas region (North and South America) are
based on a subset of data used for Section 3 and are subject to the same thresholds
and filters discussed within that section. The subset used for this section includes
connections identified as coming from networks in North and South America, based
on classification by Akamai’s EdgeScape geolocation tool.

Aside from more moderate year-over-year changes seen in Chile
(up 3.7%) and Argentina (up 8.1%), yearly growth among the
other Americas countries was more aggressive. Three countries
(Uruguay, Ecuador, and Bolivia) all grew more than 50% from
the first quarter of 2012, while an additional six grew more
than 25%. While the United States and Canada had some of
the fastest average peak connection speeds, the year-over-year
changes seen there were in the middle of the pack. Looking
into the future, it is likely that the average peak speeds seen in
a number of South American countries will continue to see ag-
gressive growth. A February 2013 blog post31
from BuddeComm,
a telecommunications research consultancy, noted that:
• Fiber-optic backbone networks are being deployed in Brazil’s
major cities as part of a National Broadband Plan
• Colombia’s National Fiber Optic Project aims to invest the
equivalent of $600 million USD to deploy over 15,000 km of
optical fiber
• Peru’s national broadband plan aims to provide Internet
connectivity to the more remote regions of the country via a
fiber-optic backbone
5.3 Americas High Broadband Connectivity
Though the first quarter average and average peak connection
speeds observed across Americas countries were respectable,
the distribution of high speed Internet connections is much
more limited, as evidenced by Figure 21. As it shows, only six
countries in North and South America had more than 25,000
unique IP addresses connecting to Akamai at speeds above 10
Mbps, qualifying them for inclusion within this section. Un-
surprisingly, the United States and Canada were well ahead of
the other qualifying countries, with high broadband adoption
rates more than 20-25 times higher than Chile, which had the
next highest adoption rate, at 0.9%. The other three quali-
fying countries (Mexico, Brazil, and Argentina) all had high
broadband adoption rates below 1% as well, with Argentina
the lowest, at 0.6%. Among the six qualifying countries, the
quarter-over-quarter changes were strong, ranging from 13%
in Brazil to 41% in Mexico. For the sake of completeness, data
for the remaining Americas countries that did not qualify for in-
clusion can also be found in Figure 21. In these countries, high
broadband adoption was essentially non-existent in Bolivia and
Paraguay, while reaching only 0.7% in Ecuador.
Among the qualifying countries, the year-over-year changes in
high broadband adoption levels were generally positive, and
strongly so. Increases ranged from a very respectable 38% in
Brazil to growth of 106% year-over-year in Mexico. Chile was
the only qualifying country to see a lower high broadband
adoption rate as compared to a year prior, dropping 12%.
Particularly high rates of yearly change were seen in Bolivia,
Ecuador, and especially Uruguay, but as these countries did
not qualify for inclusion, and as their adoption rates are so low,
these changes should be considered inconclusive.
YoY
Change
QoQ
Change
Q1 ’13
Peak Mbps
Country/Region
11 United States 36.6 11% 27%
18 Canada 34.2 13% 35%
56 Chile 20.4 2.9% 3.7%
60 Ecuador 19.4 14% 59%
62 Brazil 18.9 14% 25%
67 Mexico 17.5 17% 29%
72 Colombia 16.0 6.7% 12%
76 Argentina 15.5 9.3% 8.1%
80 Uruguay 15.3 9.7% 76%
82 Peru 15.2 10% 32%
97 Costa Rica 12.8 12% 31%
106 Paraguay 11.1 14% 14%
119 Bolivia 9.1 8.2% 50%
122 Venezuela 8.2 8.8% 20%
Global
Rank
Figure 20: Average Peak Connection Speed by Americas Country
YoY
Change
QoQ
Change
% Above
10 Mbps
Country/Region
8 United States 25% 14% 69%
15 Canada 19% 22% 77%
45 Chile 0.9% 17% -12%
46 Mexico 0.8% 41% 106%
47 Brazil 0.7% 13% 38%
48 Argentina 0.6% 28% 51%
– Ecuador 0.7% 36% 179%
– Costa Rica 0.6% 40% 56%
– Colombia 0.2% -20% -44%
– Peru 0.1% 1.7% 1.7%
– Venezuela 0.1% -1.9% 56%
– Uruguay 0.1% 33% 767%
– Bolivia 0.0% -13% 100%
– Paraguay 0.0% -50% -46%
Global
Rank
by Americas Country

5.4 Americas Broadband Connectivity
The story for broadband adoption within the surveyed Americas
countries in the first quarter of 2013 was somewhat better than
high broadband adoption, with nine of the 14 surveyed coun-
tries qualifying for inclusion, as shown in Figure 22. Consistent
with the other metrics reviewed in this section, the United States
and Canada had broadband adoption levels far and away above
those seen in the other countries. However, for broadband adop-
tion, Canada’s 77% adoption rate is 10% higher than the 70%
adoption rate seen in the United States. Both countries had nearly
identical quarterly increases as well, with Canada adding 3.9%
and the United States growing 3.8%. Although the data is from
October 2012, a report32
published by the United States National
Telecommunications Information Administration (NTIA) noted
that 72.4% of American households have high-speed Internet at
home. While not in perfect agreement, this figure is relatively close
to the broadband adoption rate observed by Akamai.
Among the other qualifying countries, broadband adoption levels
ranged from 20% in Mexico to just 1.2% in Venezuela. The high-
est and lowest quarterly changes were also seen in those same
countries, with changes ranging from 3.6% in Venezuela to 37%
in Mexico. Broadband adoption rates and insight into quarterly/
yearly changes for the five countries that did not qualify for inclu-
sion appear within Figure 22 for the sake of completeness.
Yearly changes among the qualifying Americas countries were very
mixed. Two countries, Chile and Argentina, saw broadband adop-
tion rates decline year-over-year, losing 16% and 11% respectively.
In contrast, Ecuador and Venezuela both saw extremely aggres-
sive yearly growth, gaining 218% and 98% respectively. Mexico
added 68% from the first quarter of 2012, and the remaining four
countries all posted yearly increases below 25%.
Although Cuba is not among the surveyed countries in the Ameri-
cas region, international Internet connectivity showed signs of
improvement on the island nation during the first quarter. A blog
post33
from Internet monitoring firm Renesys described signs of
change observed in Cuba’s Internet connectivity, as seen through
new routes and lower latencies, pointing to the operationalization
of the ALBA-1 submarine cable. This observation was later con-
firmed in a statement34
from the Cuban state telecommunications
company (ETECSA).
SECTION 5:
Geography –Americas (continued)
YoY
Change
QoQ
Change
% Above
4 Mbps
Country/Region
8 Canada 77% 3.9% 13%
18 United States 70% 3.8% 16%
57 Mexico 20% 37% 68%
58 Chile 18% 23% -16%
60 Brazil 14% 11% 18%
61 Ecuador 13% 13% 218%
62 Colombia 13% 11% 22%
65 Argentina 10% 13% -11%
76 Venezuela 1.2% 3.6% 98%
– Costa Rica 3.3% 10% 28%
– Peru 1.9% 22% 99%
– Uruguay 1.2% 61% 598%
– Bolivia 0.5% 5.0% 153%
– Paraguay 0.4% 32% -47%
Global
Rank
Figure 22: Broadband (4 Mbps) Connectivity by Americas Country
DID YOU KNOW?
According to a study by watchdog organization
Freedom House, Cubans can legally access the
Internet only through government-approved
institutions, such as points of access run by the
state telecommunications company ETECSA,
and users are generally required to present
identification to use computers at these sites.
[Source: http://www.freedomhouse.org/report/freedom-
net/2012/cuba]

The metrics presented here for the Asia Pacific region are based on a subset of data
used for Section 3 and are subject to the same thresholds and filters discussed
within that section.The subset used for this section includes connections identified as
coming from networks in the Asia Pacific region, based on classification by Akamai’s
EdgeScape geolocation tool.
6.1 Asia Pacific Average Connection Speeds
In the first quarter of 2013, three Asia Pacific countries/regions
saw average connection speeds above the “high broadband”
threshold of 10 Mbps. Not only were they the top three Asia
Pacific countries/regions, but they topped the global ranks as
well. As shown in Figure 23, South Korea, Japan, and Hong
Kong all saw modest quarterly increases, growing to 14.2
Mbps, 11.7 Mbps, and 10.9 Mbps respectively. Average con-
nection speeds in the remaining Asia Pacific countries/regions
ranged from 6.9 Mbps in Singapore down to 1.3 Mbps in India.
Quarter-over-quarter changes across these countries/regions
were generally positive, with only China and Vietnam seeing
minor quarterly losses. Among those seeing quarterly increases,
the gains were as high as 10% in Singapore and 20% in India.
Looking at year-over-year changes in the Asia Pacific region, we
find that they were generally positive in the first quarter, and
were fairly strong as well. Japan was the only country in the
region to see a yearly increase below 10% (at 6.8%), while In-
donesia saw the most significant growth, up 113% from a year
prior. Four additional countries grew more than 30% year-over-
year, and three more saw increases in the 20% range. Taiwan,
SECTION 6:
Geography –Asia Pacific Region
YoY
Change
QoQ
Change
Q1 ’13
Avg. Mbps
Country/Region
1 South Korea 14.2 1.5% -10%
2 Japan 11.7 3.9% 6.8%
3 Hong Kong 10.9 9.0% 16%
21 Singapore 6.9 10% 31%
41 Australia 4.7 2.1% 31%
43 New Zealand 4.4 6.2% 14%
44 Thailand 4.4 2.4% 32%
47 Taiwan 4.3 4.8% 10%
70 Malaysia 2.7 9.2% 36%
98 China 1.7 -5.6% 20%
104 Indonesia 1.5 6.9% 113%
108 Vietnam 1.5 -3.1% -9.1%
109 Philippines 1.4 3.7% 24%
114 India 1.3 20% 21%
Global
Rank
Figure 23: Average Connection Speed by
Asia Pacific Country/Region
DID YOU KNOW?
• According to the 2013 ICT Facts Figures update from the International Telecommunications
Union, 32% of people in the Asia Pacific region use the Internet, while an estimated 33% of
households in the Asia Pacific region have Internet access.
[Source: http://www.itu.int/en/ITU-D/Statistics/Documents/facts/ICTFactsFigures2013.pdf]
• In March, Telecom New Zealand launched its “next generation broadband” plans for services on
the government-funded Ultra-Fast Broadband (UFB) fiber network. The UFB network offer two
tiers of connection speeds: 30 Mbps down, 10 Mbps up and 100 Mbps down, 50 Mbps up.
[Source: http://www.zdnet.com/au/telecom-nz-announces-ultra-fast-broadband-plans-7000013162/]
New Zealand, and Hong Kong rounded out the group with
year-over-year growth at or above 10%. South Korea and Viet-
nam were the only two Asia Pacific countries to see negative
year-over-year changes, declining 10% and 9.1% respectively.

SECTION 6:
Geography –Asia Pacific Region (Continued)
6.2 Asia Pacific Average Peak Connection Speeds
As shown in Figure 24, Hong Kong remained far and away the
top region with the highest average peak connection speed,
topping both the Asia Pacific and global ranks at 63.6 Mbps,
up 9.0% quarter-over-quarter. Japan was the only other Asia
Pacific country that had an average peak connection speed at/
above 50 Mbps, after a strong 13% quarterly increase. Both
South Korea and Singapore posted average peak connec-
tion speeds above 40 Mbps, while Taiwan and Thailand both
leverage quarterly gains in the 6% range to achieve average
peak speeds above 30 Mbps. The remaining surveyed countries,
with the exception of China, all had average peak connection
speeds above 10 Mbps in the first quarter. Up 4.1% quarter-
over-quarter, China had the lowest average peak connection
speed among the surveyed Asia Pacific countries/regions, at 8.3
Mbps. Quarterly changes among the surveyed countries/regions
ranged from a low of 1.5% in South Korea to 27% in India;
Japan, Malaysia, and New Zealand also saw quarterly growth in
excess of 10%.
Year-over-year changes across the surveyed Asia Pacific coun-
tries/regions were particularly strong in the first quarter of 2013.
Indonesia posted the greatest gain, at 152%, while Japan, Tai-
wan, and New Zealand all had the lowest level of yearly growth,
at a still impressive 26%. Three countries (Thailand, Australia,
and Malaysia) all grew by more than 50% year-over-year, with
Australia’s gain more than erasing the unexpected yearly loss
seen in the fourth quarter of 2012. China and Singapore both
grew more than 40% year-over-year, while the Philippines, In-
dia, and Vietnam all added more than 30%. With the exception
of South Korea, the remaining countries/regions all saw yearly
increases in excess of 20%.
6.3 Asia Pacific High Broadband Connectivity
In the first quarter of 2013, there was a very broad range of high
broadband adoption rates across the Asia Pacific countries/re-
gions that qualified for inclusion. South Korea saw half of its con-
nections to Akamai at speeds above 10 Mbps, while China saw
just one fifth of one percent of connections to Akamai at those
speeds. As shown in Figure 25, Japan, Hong Kong, and Singa-
pore joined South Korea as the only other Asia Pacific countries/
regions with high broadband adoption levels above 10%. The
remaining countries all had adoption rates below 5%, ranging
from 4.8% in Australia down to China’s 0.2%. Quarterly changes
among the qualifying countries were somewhat mixed, with
increases ranging from just 0.8% in Australia to a massive 102%
in India. Thailand and China saw declines of 0.5% and 21%, re-
spectively. The Philippines, Indonesia, and Vietnam did not qualify
for inclusion in the global ranking, but data from these countries
is included in Figure 25 for the sake of completeness. As shown,
high broadband adoption in these countries is extremely low.
YoY
Change
QoQ
Change
Q1 ’13
Peak Mbps
Country/Region
1 Hong Kong 63.6 9.0% 29%
2 Japan 50.0 13% 26%
4 South Korea 44.8 1.5% -6.3%
6 Singapore 41.1 8.8% 43%
21 Taiwan 31.3 6.8% 26%
25 Thailand 30.1 6.3% 56%
37 Australia 26.3 9.3% 58%
46 Malaysia 23.6 11% 53%
57 New Zealand 20.2 10% 26%
88 Philippines 13.9 3.6% 39%
99 Indonesia 12.8 -0.4% 152%
105 Vietnam 11.6 7.3% 36%
109 India 10.6 27% 34%
121 China 8.3 4.1% 41%
Global
Rank
Figure 24: Average Peak Connection Speed
by Asia Pacific Country/Region
YoY
Change
QoQ
Change
% Above
10 Mbps
Country/Region
1 South Korea 50% 1.5% -6.2%
2 Japan 43% 5.7% 17%
3 Hong Kong 34% 10% 20%
18 Singapore 16% 20% 101%
36 Australia 4.8% 0.8% 34%
37 Taiwan 4.5% 12% 53%
38 New Zealand 3.7% 28% 55%
41 Thailand 2.7% -0.5% 196%
43 Malaysia 1.5% 24% 116%
49 India 0.3% 102% 85%
50 China 0.2% -21% 62%
– Philippines 0.2% 25% 36%
– Indonesia 0.2% -22% 94%
– Vietnam 0.1% 18% -20%
Global
Rank

On a year-over-year basis, high broadband adoption rates in the
qualifying countries/regions generally saw significant improve-
ments. Singapore, Malaysia, and Thailand all more than doubled
adoption rates as compared to the first quarter of 2012, increas-
ing 101%, 116%, and 196% respectively. India, China, New Zea-
land, and Taiwan also saw strong growth, posting yearly increases
of over 50%. Among the three countries that did not qualify for
inclusion in the global rankings, Indonesia and the Philippines saw
solid yearly growth, while Vietnam saw a year-over-year loss.
6.4 Asia Pacific Broadband Connectivity
As Figure 26 shows, all but one of the surveyed Asia Pacific
countries/regions qualified for inclusion in the global rankings in
the first quarter of 2013 — the Philippines was the lone exclu-
sion. Among the qualifying countries/regions, the spread of
adoption rates was significantly greater than that seen for the
high broadband adoption metric, ranging from 87% in South
Korea all the way down to 1.9% in Vietnam. Adoption rates
continued to be fairly strong in most countries/regions, though
China, Indonesia, India, and Vietnam maintained adoption rates
below 10%. Quarter-over-quarter changes were mostly posi-
tive, and fairly nominal, across the qualifying countries/regions.
While Malaysia and India did show quarterly increases above
10% (growing 18% and 33% respectively), growth in the other
countries/regions fell below that level. Australia, China, and
Vietnam all posted quarterly declines in broadband adoption,
losing 5.6%, 12%, and 1.3% respectively.
Year-over-year changes were a mixed bag as well, with a broad
range of growth rates, as well as losses seen in two qualifying
countries/regions. An unusually high year-over-year change of
414% was seen in Indonesia, while Thailand, Malaysia, and
China all posted growth rates just over 100%. Yearly growth
YoY
Change
QoQ
Change
% Above
4 Mbps
Country/Region
2 South Korea 87% 1.5% 1.5%
5 Japan 79% 3.8% 9.9%
6 Hong Kong 78% 3.3% 8.4%
25 Singapore 62% 9.4% 22%
38 Thailand 48% 6.2% 108%
42 New Zealand 42% 8.7% 22%
44 Australia 37% -5.6% 38%
47 Taiwan 33% 2.1% -8.7%
54 Malaysia 21% 18% 103%
71 China 5.3% -12% 101%
72 Indonesia 3.7% 2.7% 414%
74 India 2.4% 33% 66%
75 Vietnam 1.9% -1.3% -48%
– Philippines 1.8% 18% 46%
Global
Rank
Figure 26: Broadband (4 Mbps) Connectivity
below 10% was seen in the top three ranked countries/regions,
with South Korea growing 1.5%, Japan growing 9.9%, and
Hong Kong growing 8.4%. Although China’s broadband adop-
tion rate doubled on year-over-year basis, the research chief of
China’s Ministry of Industry and Information Technology indi-
cated that there is still a significant gap between China’s broad-
band coverage and that of developed nations, and that the lag
is increasing.35
While the Chinese government has said that it
aims to increase broadband coverage to 95% percent by 2015,
challenges to driving significant ongoing improvements to
adoption rates include the relatively high price of broadband,
poor connections and customer service, as well as disparity in
coverage between cities and rural areas, and a general lack of
infrastructure.
DID YOU KNOW?
Many Chinese Internet users live in residential communities with only one broadband provider, but un-
der new regulations announced in January by China’s Ministry of Industry and Information Technology
(MIIT), residential communities with fiber-optic broadband cables must allow multiple telecommunica-
tions operators access to give residents a choice in all county-size and larger cities across the country.
[Source: http://www.techinasia.com/chinese-broadband-customers-finally-choose-providers/]

The metrics presented here for the EMEA region are based on a subset of data used
for Section 3 and are subject to the same thresholds and filters discussed within that
section. The subset used for this section includes connections identified as coming
from networks in the EMEA region, based on classification by Akamai’s EdgeScape
geolocation tool.
7.1 EMEA Average Connection Speeds
After topping the list during 2012, Switzerland remained the
EMEA country with the highest average connection speed in
the first quarter of 2013. As shown in Figure 27, Switzerland’s
10.1 Mbps “high broadband”-level average connection speed
was the fastest in the region, though it was once again followed
closely by the Netherlands, which grew 10% quarter-over-
quarter to 9.9 Mbps, just below the high broadband threshold.
The greatest concentration of average connection speeds was
between 6-8 Mbps, with 14 of the surveyed countries falling
within that range. Quarter-over-quarter changes among the
EMEA countries were all positive in the first quarter, ranging
from 3.3% in Romania to 14% in Austria. In addition to Austria,
the Netherlands, Sweden, Denmark, Ireland, and Turkey also
saw quarterly growth in excess of 10%. South Africa remained
the EMEA country with the lowest average connection speed,
falling to 1 Mbps behind Turkey, at 2.1 Mbps after a 3.5% quar-
terly increase. These two countries were also the only two within
the EMEA region that had average connection speeds below the
“broadband” threshold.
With the exception of twin 1.5% year-over-year declines seen
in both Ireland and Portugal, yearly changes among surveyed
EMEA countries were fairly positive in the first quarter. Israel had
the highest rate of change, at 51%, while the United Kingdom
also saw aggressive yearly growth, at 41%. Five additional coun-
tries improved 30% or more over the prior year, while another
three grew 20% or more. The lowest rate of change was seen
in Belgium, which increased just 3.6% from the first quarter
of 2012. Although it had the lowest average connection speed,
South Africa turned in a respectable yearly increase, at 16%.
7.2 EMEA Average Peak Connection Speeds
Romania’s 47.9 Mbps average peak connection speed once again
placed it solidly ahead of Switzerland, which was the next fast-
est EMEA country in the first quarter of 2013, with an average
peak connection speed of 40.3 Mbps. As is shown in Figure 28,
Romania and Switzerland were the only two surveyed countries in
the EMEA region to have average peak connection speeds above
SECTION 7:
Geography – Europe/Middle East/Africa (EMEA)
YoY
Change
QoQ
Change
Country/Region
Global
Rank
Q1 ‘13
Avg. Mbps
Figure 27: Average Connection Speed by EMEA Country
4 Switzerland 10.1 6.1% 24%
7 Czech Republic 9.6 9.1% 34%
8 Sweden 8.9 10.8% 32%
10 Denmark 8.2 13% 17%
11 Austria 7.9 14% 14%
12 United Kingdom 7.9 7.3% 41%
14 Finland 7.7 5.5% 13%
15 Romania 7.5 3.3% 13%
16 Norway 7.4 6.5% 30%
17 Belgium 7.4 7.1% 3.6%
18 Ireland 7.3 12% -1.5%
19 Israel 7.0 6.4% 51%
20 Germany 6.9 8.4% 20%
24 Hungary 6.6 5.9% 12%
25 Slovakia 6.4 6.6% 12%
27 Poland 6.2 8.6% 25%
28 United Arab Emirates 6.2 8.8% 32%
29 Russia 6.0 8.4% 32%
35 Portugal 5.3 3.5% -1.5%
36 Spain 5.2 4.3% 14%
38 France 5.2 5.6% 5.2%
45 Italy 4.4 4.4% 5.4%
61 Turkey 3.1 12% 9.6%
80 South Africa 2.1 3.5% 16%
40 Mbps in the first quarter. More than half of the countries (13)
saw average peak connection speeds above 30 Mbps, while all
but one of the remaining EMEA countries were above 20 Mbps
in the first quarter. The lone standout was South Africa, which
maintained its position as the only one of the surveyed countries
below 10 Mbps, with an average peak connection speed that
increased 12% quarter-over-quarter to 7.6 Mbps. In addition to
South Africa, 17 other countries saw quarterly growth of 10% or
more, with Turkey’s 21% the highest in the region. The lowest
rate of growth was found in Portugal, which added 6.5% from
the fourth quarter of 2012.

Looking at year-over-year changes, growth was particularly strong
across all of the surveyed EMEA countries. Impressively, the lowest
level of yearly growth was 20%, seen in Slovakia. Nine additional
surveyed countries saw yearly increases in the 20% range, seven
more in the 30% range, five in the 40% range, while the United
Kingdom and Israel saw the largest rates of year-over-year growth,
at 53% and 52% respectively.
This aggressive growth may continue into the future as well, if
proposals made by the European Commission in March are suc-
cessfully implemented. In a speech36
made by European Com-
mission Vice President Neelie Kroes, four EU-wide actions were
proposed with the aim of “Getting every home and town in
Europe ‘broadband ready’”:
1. Ensuring that telecommunications companies know what physi-
cal infrastructure is already in place, and opening up access to it
with fair and reasonable prices and conditions
2. Increasing coordination for “new works,” with the goal of mak-
ing it easier for operators and providers to work together
3. Cutting the complexity and cost associated with the red tape as-
sociated with getting permits for broadband and mobile buildout
4. Ensuring “broadband-ready buildings,” with all new buildings
and major renovations equipped for high-speed broadband
7.3 EMEA High Broadband Connectivity
Similar to the average connection speed metric, Switzerland
just barely beat out the Netherlands to take the top spot for
the highest level of high broadband adoption among surveyed
EMEA countries in the first quarter of 2013. Figure 29 shows
that 30% of connections to Akamai from Switzerland in the
first quarter were at speeds above 10 Mbps, while the Neth-
erlands was just 1% behind. At least a fifth of connections to
Akamai from six additional countries were at 10 Mbps or above,
while 10 more countries had high broadband adoption rates
above 10%. Turkey was once again the EMEA country with the
lowest level of high broadband adoption, but interestingly, was
also the one with the largest quarterly increase, growing 68%
quarter-over-quarter to 1.0%. In contrast, the smallest quarterly
increases were seen in neighboring Spain and Portugal, which
grew 4.1% and 4.6% respectively. Among the other surveyed
countries in the EMEA region, quarter-over-quarter changes
were generally fairly strong, with five countries (not including
South Africa) growing more than 20%, and another dozen see-
ing quarterly increases of 10% or above.
YoY
Change
QoQ
Change
Country/Region
Global
Rank
3 Romania 47.9 8.9% 23%
7 Switzerland 40.3 12% 41%
10 Belgium 38.0 14% 30%
12 United Kingdom 36.3 12% 53%
13 Hungary 35.9 12% 28%
14 Israel 35.9 10% 52%
15 Czech Republic 35.5 11% 45%
16 Sweden 34.9 19% 38%
17 Portugal 34.5 6.5% 22%
19 Poland 32.2 17% 46%
20 Spain 31.3 14% 31%
22 Ireland 30.9 15% 22%
23 Germany 30.8 12% 31%
24 Austria 30.6 16% 29%
27 Denmark 29.5 15% 32%
29 Russia 29.3 18% 43%
30 Finland 29.2 9.5% 24%
31 Slovakia 29.1 9.5% 20%
32 Norway 28.7 14% 43%
42 Turkey 25.0 21% 35%
47 France 23.5 9.9% 23%
50 Italy 21.8 9.7% 24%
126 South Africa 7.6 12% 25%
Q1 ‘13
Peak Mbps
Figure 28: Average Peak Connection Speed by EMEA Country
Year-over-year increases above 100% were surprisingly fre-
quent in the first quarter, with the United Kingdom posting
a 208% yearly gain, while the United Arab Emirates grew
129%, Turkey 130%, South Africa 126%, and Israel 117%.
Strong yearly growth was also seen in Switzerland, Sweden,
the Czech Republic, Norway, Germany, Poland, and Russia,
which all added 50% or more year-over-year. Belgium saw
the smallest year-over-year change among surveyed EMEA
countries, up just 2.1%, and Portugal was the only EMEA
country to see a yearly decline in its high broadband adoption
rate, losing 11%.
The United Kingdom’s impressive yearly gain can be seen as an
indicator of strong improvements in the state of Internet con-
nectivity within the country. In some communities, residents
are getting access to gigabit-speed connections. One example
is in Oxfordshire, where a privately funded company known as
Gigaclear announced37
in February that it would start provid-
ing 1 Gbps connections to local businesses and residents.
Another is Arkholme, a village in rural Lancashire, where local

residents implemented a DIY effort, digging their own trenches
and laying their own fiber, bring 1 Gbps FTTH connectivity to
over 150 homes by February 2013.38
7.4 EMEA Broadband Connectivity
Broadband adoption across surveyed EMEA countries remained
extremely strong in the first quarter of 2013, as evidenced
by Figure 30. Switzerland led both the EMEA region and the
global rankings by having the highest broadband adoption rate,
with 88% of the connections to Akamai from the country at
speeds above 4 Mbps. The Netherlands and the Czech Republic
joined Switzerland in having broadband adoption rates above
80%, coming in at 84% and 81% respectively. Impressively,
another eight of the surveyed EMEA countries saw broadband
adoption rates grow to, or remain at, more than 70%. With
the exception of Italy, Turkey, and South Africa, the remain-
ing countries also had more than half of their connections to
Akamai at speeds above 4 Mbps. South Africa was the only
surveyed EMEA country with a broadband adoption rate below
10% in the first quarter, at 8.0% adoption after an 8.5%
SECTION 7:
Geography – Europe/Middle East/Africa (EMEA) (Cont.)
YoY
Change
QoQ
Change
Country/Region
Global
Rank
% Above
10 Mbps
by EMEA Country
5 Switzerland 30% 11% 58%
6 Netherlands 29% 25% 21%
7 Sweden 25% 18% 61%
9 Czech Republic 23% 10% 96%
10 Finland 21% 12% 30%
11 United Kingdom 20% 16% 208%
12 Norway 20% 6.7% 66%
13 Denmark 20% 21% 21%
14 Belgium 19% 19% 2.1%
16 Romania 19% 5.7% 39%
19 Austria 15% 19% 9.8%
22 United Arab Emirates 13% 39% 129%
23 Germany 13% 19% 53%
24 Poland 13% 18% 80%
25 Israel 12% 5.9% 117%
26 Russia 12% 23% 84%
27 Hungary 12% 14% 18%
28 Ireland 12% 27% 13%
31 Slovakia 9.6% 11% 7.9%
33 Portugal 6.0% 4.6% -11%
34 Spain 5.8% 4.1% 37%
35 France 5.2% 5.8% 16%
39 Italy 3.2% 7.8% 24%
42 South Africa 1.5% 10% 126%
44 Turkey 1.0% 68% 130%
YoY
Change
QoQ
Change
Country/Region
Global
Rank
% Above
4 Mbps
Figure 30: Broadband (4 Mbps) Connectivity by EMEA Country
1 Switzerland 88% 2.5% 14%
3 Netherlands 84% 1.2% 1.2%
4 Czech Republic 81% 8.7% 19%
7 Austria 78% 16% 28%
9 Denmark 77% 11% 14%
10 Belgium 76% 4.5% 5.1%
11 Hungary 75% 7.5% 23%
13 Romania 73% 4.4% 15%
14 Israel 73% 9.0% 82%
15 United Kingdom 73% 4.5% 25%
16 Germany 72% 8.6% 26%
20 Finland 66% 4.7% 11%
22 Russia 63% 11% 47%
23 Sweden 63% 15% 20%
24 United Arab Emirates 62% 5.5% 41%
27 Portugal 61% 7.0% -0.4%
28 Poland 58% 11% 33%
30 Ireland 56% 8.7% 8.1%
32 Spain 55% 10% 23%
33 France 53% 11% 7.8%
35 Slovakia 52% 4.4% 30%
37 Norway 51% 4.9% 13%
45 Italy 35% 12% 10%
59 Turkey 16% 46% 35%
69 South Africa 8.0% 8.5% 16%
quarterly increase. However, South Africa’s quarterly increase
was more-or-less middle of the pack, with Turkey’s 46% growth
the highest, and the Netherlands’ 1.2% the lowest, with 14
other countries (more than half of those surveyed) also seeing
quarter-over-quarter changes below 10%.
Similar to the high broadband metric, Portugal was the only sur-
veyed EMEA country to see a year-over-year decline in broadband
adoption, though it was very slight, at just 0.4%. Within the EMEA
region, Israel had the largest year-over-year change, adding 82%
from the same period a year prior. An additional five countries grew
30% or more, while 14 more were up 10% or more and just four
countries saw yearly increases below 10%. While clearly not a driver
of the extremely high year-over-year growth seen in Israel in the
first quarter, the country’s planned national high-speed broadband
network could help drive similar long-term growth in the future.
In January, Israel’s state-owned electric company announced plans,
with government backing, to roll out FTTH-based gigabit connec-
tivity; it aims to have 10% of the country wired by next year, and
two-thirds of the country covered within seven years.39

As has been noted in prior quarters, the source data set for this
section is subject to the following constraints:
• A minimum of 1,000 unique IP addresses connecting to
Akamai from the network in the first quarter of 2013 was
required for inclusion in the list.
• In countries where Akamai had data for multiple network
providers, only the top three are listed, based on unique IP
address count.
• The names of specific mobile network providers have been
made anonymous, and providers are identified by a unique ID.
• Data is included only for networks where Akamai believes
that the entire Autonomous System (AS) is mobile — that is,
if a network provider mixes traffic from fixed/wireline (DSL,
cable, etc.) connections with traffic from mobile connections
on a single network identifier, that AS was not included in
the source data set.
• Akamai’s EdgeScape database was used for the geographic
assignments.
8.1 Connection Speeds on Mobile Networks
In the first quarter of 2013, Russian mobile provider RU-1
returned to its position as the provider with the highest aver-
age connection speed, at 8.6 Mbps, thanks to a 7.3% quarterly
increase. It bested second-place provider AT-2 by 1 Mbps, which
lost 5.9% quarter-over-quarter to 7.6 Mbps, after the two had
swapped places in the fourth quarter of 2012. In reviewing the
full list of mobile providers shown in Figure 31, we find that
there are ten providers (CZ-3, GR-1, UA-1, DE-2, RU-4, AT-1, NO-
1, CA-2, US-2, and ES-1) that had average connection speeds in
the “broadband” (4 Mbps) range. Similar to last quarter, an ad-
ditional 63 providers had average connection speeds greater than
1 Mbps, including provider SV-3 in El Salvador, which was just 6
kbps over the threshold, boosted by a 20% quarterly increase.
This increase was among the top 20 largest seen, with quarterly
growth of 47% (RU-4, to 4.8 Mbps) and 41% (BO-2, to 0.9
Mbps) topping the list. The smallest quarterly increase was seen
at provider TW-1, which added 0.6% to 1.7 Mbps. Quarterly
declines were seen at 24 providers, ranging from less than 0.1%
at US-3 to 17% at UY-1. Nigeria’s NG-1 remained the mobile
provider with the lowest average connection speed, at 0.4 Mbps
(up 49 kbps from the fourth quarter of 2012). Including NG-1, a
total of nine providers had average connection speeds below 1
Mbps in the first quarter.
Examining the average peak connection speed data for the
first quarter of 2013, we find that Hong Kong provider HK-1
vaulted to the top of the list at 45.6 Mbps, thanks to a 50% in-
crease from the prior quarter. Last quarter’s leader ES-1 dropped
further down the list due to an unusually high 36% quarterly
decline to 28.0 Mbps. RU-1, which led the average connec-
tion speed rankings, placed second for this metric, with a 43.7
Mbps average peak connection speed and an 8.0% quarter-
over-quarter change. These two were the only providers with
average peak connection speeds over 40 Mbps, and only three
more (DE-2, SK-2, AT-2) had speeds over 30 Mbps. However, an
additional 13 providers posted average peak connection speeds
over 20 Mbps, while another 39 were above 10 Mbps. The
lowest average peak connection speed in the first quarter, 2.8
Mbps, was seen in South Africa, on provider ZA-1, up 2.8%
quarter-over-quarter. This increase was one of the smallest
seen in the first quarter, with changes ranging from 1.4% at
provider IE-3 to SA-1’s jump of 123%, and TW-2’s 130% leap.
Thirteen mobile providers saw average peak connection speeds
decline as compared to the prior quarter, with losses ranging
from 0.7% at provider DE-1 to the previously mentioned 36%
decline at Spanish mobile provider ES-1.
Year-over-year trends for the average connection speed metric
were, by and large, fairly encouraging in the first quarter. Mobile
provider US-2 improved by nearly 300% from the first quarter
of 2012, while Slovakian provider SK-1 was up nearly 200%. An
additional 11 providers saw average connection speeds grow by
50% or more year-over-year, while another 23 were up 20% or
more over the same period. Twenty-one providers did see aver-
age connection speeds decline on a yearly basis, with losses rang-
SECTION 8:
Mobile Connectivity
The source data in this section encompasses usage not only from smartphones, but
also laptops, tablets, and other devices that connect to the Internet through these
mobile networks. In addition, this edition of the State of the Internet Report once
again includes insight into mobile traffic growth and data traffic patterns contributed
by Ericsson, a leading provider of telecommunications equipment and related services
to mobile and fixed network operators globally.

ASIA
Q1 ‘13
Peak
Mbps
Q1 ‘13
Avg.
Mbps
ID
Q1 ‘13
Peak
Mbps
Country
AFRICA
Q1 ‘13
Avg.
Mbps
IDCountry
Figure 31: Average and Average Peak Connection Speeds by Mobile Provider
Egypt EG-1 0.8 6.1
Morocco MA-1 1.5 15.8
Nigeria NG-1 0.4 5.3
South Africa ZA-1 0.5 2.8
ASIA
China CN-1 2.2 6.5
Hong Kong HK-2 2.9 17.1
Hong Kong HK-1 2.9 45.6
Indonesia ID-1 0.8 11.9
Kuwait KW-1 1.4 12.2
Malaysia MY-3 1.6 10.6
Pakistan PK-1 1.5 10.3
Qatar QA-1 1.4 12.3
Saudi Arabia SA-1 1.5 12.6
Singapore SG-3 2.2 12.0
Sri Lanka LK-1 1.4 13.7
Taiwan TW-1 1.7 12.4
Taiwan TW-2 1.5 12.8
Thailand TH-1 1.1 26.9
EUROPE
Austria AT-1 4.8 22.3
Austria AT-2 7.6 32.4
Belgium BE-3 1.7 11.6
Belgium BE-2 2.1 6.8
Czech Republic CZ-3 6.4 20.4
Estonia EE-1 1.6 8.6
France FR-2 2.8 11.1
Germany DE-1 1.3 6.2
Germany DE-2 5.7 32.8
Greece GR-1 6.1 21.7
Hungary HU-2 2.3 12.1
Hungary HU-1 1.7 11.6
Ireland IE-1 3.3 18.5
Israel IL-1 1.7 9.1
Italy IT-2 3.1 19.3
Italy IT-3 3.7 19.6
Italy IT-4 2.2 17.4
Lithuania LT-2 2.5 21.5
Moldova MD-1 2.8 12.8
Netherlands NL-2 1.9 6.7
Netherlands NL-1 2.3 6.1
Norway NO-1 4.6 29.2
Poland PL-1 3.1 19.0
Romania RO-1 1.1 7.4
Russia RU-1 8.6 43.7
Russia RU-4 4.8 26.0
Russia RU-3 1.0 8.1
Slovakia SK-1 1.7 9.1
Slovakia SK-2 3.1 32.7
Slovenia SI-1 2.4 9.8
Spain ES-1 4.1 28.0
Turkey TR-1 2.0 13.6
Ukraine UA-1 6.0 21.6
United Kingdom UK-3 3.0 17.3
NORTH AMERICA
Canada CA-2 4.3 12.9
El Salvador SV-2 1.8 13.3
Guatemala GT-2 2.6 14.1
United States US-2 4.3 15.6
OCEANIA
Australia AU-3 2.4 15.9
New Caledonia NC-1 0.7 5.7
New Zealand NZ-2 2.0 13.4
SOUTH AMERICA
Argentina AR-1 0.9 5.1
Argentina AR-2 2.3 24.4
Bolivia BO-1 0.9 5.5
Brazil BR-1 0.9 8.8
Brazil BR-2 1.3 11.6
Chile CL-3 2.0 19.5
Chile CL-4 1.5 15.9
Colombia CO-1 1.4 9.5
Paraguay PY-2 1.3 9.0
Uruguay UY-1 1.5 12.4
Venezuela VE-1 1.1 8.5
NORTH AMERICA
OCEANIA
SOUTH AMERICA
ASIA
EUROPE
SECTION 8:
Mobile Connectivity (Continued)

ing from 1.2% at provider UK-1 to nearly 18% at Czech provider
CZ-1. The long-term story is even better, as expected, when we
look at the average peak connection speed metrics. Four mobile
providers (QA-1, US-2, NO-1, TH-1) grew by more than 200%
year-over-year, while four more (SA-1, BE-3, SK-1, SK-2) added
over 100%. Another 17 providers saw average peak connec-
tion speeds increase by 50% or more, while 31 more grew by
20% or more over the past year. Just seven mobile providers saw
average peak connection speeds decline on a yearly basis, losing
anywhere from 2.7% (HU-2) to nearly 49% (MY-2).
8.2 Mobile Browser Usage Data
In June 2012, Akamai launched the “Akamai IO” destination
site (http://www.akamai.com/io), with an initial data set that
highlights browser usage across PC and mobile devices, con-
necting via fixed and mobile networks. Note that the data set
used for the figures in this section has historically come from
sampling traffic across several hundred top-tier sites deliver-
ing content through Akamai, most of which are focused on a
U.S. audience, biasing the data presented in favor of U.S. users.
Planned back-end data collection and processing improvements
should expand the sample set, which will allow us to provide
more global and geo-specific views of the data, as well as more
granular insight into browser versions. An initial release of this
updated data source occurred in mid-February 2013, which
drove some significant changes in adoption levels, as shown be-
low. Future updates will allow us to provide insight into browser
version trends as well as geo-specific trends in upcoming issues
of the State of the Internet Report.
Figure 32 illustrates browser usage by users identified to be on
cellular networks in the first quarter of 2013.40
As noted below,
an update to the data source was made in mid-February, and
is evident by large shifts in the lines plotted in the figure. This
update also drove the decision to focus on just Android Webkit,
Apple Mobile Safari, and “Others” for the first quarter overview.
As is shown in the graph, for the first half of the quarter, both
of the named browsers were relatively close in usage, similar
to what has been observed in prior quarters. Android Webkit
trended to an average of just over 41% of requests, while
Apple Mobile Safari saw nearly 38% of requests.41
However,
the gap between the two browsers widened considerably in the
second half of the quarter, with Android Webkit responsible for
nearly 44% of requests, and Apple Mobile Safari dropping to
just over 30%.42
It is not clear what caused the unusually large
drops observed on March 25.
The impact of the updated data source is significantly different
when we examine mobile browser usage across all networks43
(not just those identified as “cellular”) for the first quarter of
2013. As Figure 33 illustrates, the updated data did not cause
much variation in the percentage of requests from Apple Mo-
bile Safari — it came at just under 60%44
for the first half of the
quarter, and just over 60%45
for the second half of the quarter.
However, Android usage rises significantly coincident with the
data update, averaging just over 20%46
for the first half of the
quarter, and just under 33% for the second half. Interestingly,
the “Others” category appears to see a decline that mirrors the
increase seen for Android Webkit — this could possibly indicate
that use of a newer version of Akamai’s device characterization
engine with the updated data set caused it to start classifying a
particular user agent as Android Webkit, where it had previ-
ously been counted as part of the Others category.
Figure 32: Mobile Browsers Seen Across Cellular Networks, Q1 2013
Android Webkit
Mobile Safari
Others
60
50
40
30
20
10
0
1/1
1/6
1/11
1/16
1/21
1/26
1/31
2/5
2/10
2/15
2/20
2/25
3/2
3/7
3/12
3/17
3/22
3/27
PercentageofRequests(%)

8.3 Mobile Traffic Growth as Observed by Ericsson
In mobile networks, the access medium (spectrum) is being
shared by different users in the same cell. It is important to
understand traffic volumes and usage patterns in order to en-
able a good customer experience. Ericsson’s presence in more
than 180 countries and its customer base representing more
than 1,000 networks enables it to measure mobile voice and
data volumes. The result is a representative base for calculat-
ing world total mobile traffic in 2G, 3G, and 4G networks (not
including DVB-H, WiFi, and Mobile WiMax).
Figure 33: Mobile Browsers Seen Across All Networks, Q1 2013
Android Webkit
Mobile Safari
Others
70
60
50
40
30
20
10
0
1/1
1/6
1/11
1/16
1/21
1/26
1/31
2/5
2/10
2/15
2/20
2/25
3/2
3/7
3/12
3/17
3/22
3/27
PercentageofRequests(%)
1600
1400
1200
1000
800
600
400
200
0
Q1
‘07
Q2
‘07
Q3
‘07
Q4
‘07
Q1
‘08
Q2
‘08
Q3
‘08
Q4
‘08
Q1
‘09
Q2
‘09
Q3
‘09
Q4
‘09
Q1
‘10
Q1
‘11
Q2
‘11
Q3
‘11
Q4
‘11
Q1
‘12
Q2
‘10
Q3
‘10
Q4
‘10
Q2
‘12
Q3
‘12
Q4
‘12
Q1
‘13
Total(UL+DL)monthlytraffic(PetaByes/month)
DataVoice
Figure 34: Total Monthly Mobile Voice and Data as Measured by Ericsson
These measurements have been performed for several years. It is
important to note that the measurements of data and voice traffic
in these networks (2G, 3G, 4G/LTE) around the world show large
differences in traffic levels between markets and regions, and also
between operators due to their different customer profiles.
As illustrated in Figure 34, the volume of mobile data traffic doubled
from the first quarter of 2012 to the first quarter of 2013, and grew
19% between the fourth quarter of 2012 and first quarter of 2013.
Note that mobile voice traffic continues to grow as well, though at a
slower rate, increasing just 4% from the first quarter of 2012 to the
first quarter of 2013.
SECTION 8:
Mobile Connectivity (Continued)

SECTION 9:
Internet Events Disruptions
9.1 Hugo Chavez Dies
Venezuelan president Hugo Chavez served as the President of
Venezuela from 1999 through his death on March 13, 2013. A
prominent and controversial leader, Chavez had been seriously ill
with cancer for more than a year prior to his passing.48
News of his
death spread quickly, with Venezuelan citizens turning to online
news sources for more information. As shown in Figure 35, Aka-
mai HTTP traffic (in hits/second) from customers classified in the
“Media Entertainment” category, delivered to users in Venezu-
ela, began to climb aggressively about an hour after Chavez died,
peaking nearly two-thirds higher than traffic levels seen just before
his death. Note that while not shown in the figure, the Mbps traf-
fic levels spike in a similar fashion as well.
On March 13, white smoke appeared at 7:06 PM local time (2:06
PM Eastern Time).51
Within minutes, live streaming traffic on the
Akamai Intelligent Platform began to climb, as shown in Figure
36. Anticipation, and traffic, grew for the next hour, until the se-
lection of Jorge Mario Bergoglio as the newly elected Pope Fran-
cis was announced at 8:15 PM local time (3:15 PM Eastern time).
Live streaming traffic continued to grow for a few more minutes
after the announcement, reaching a peak of 2197 Gbps (just
under 2.2 Tbps) at 8:25 PM local time (3:25 PM Eastern time).
Figure 35: Spike in Akamai “Media Entertainment”Traffic
to Venezuela on March 6, 2013
Hugo Chavez dies
20:55 UTC
04-Mar
00:00UTC
04-Mar
12:00UTC
05-Mar
00:00UTC
05-Mar
12:00UTC
06-Mar
00:00UTC
06-Mar
12:00UTC
Hitspersecond
Figure 36: Akamai Live Streaming Traffic on March 13, 2013
12:30
13:00
13:30
14:00
14:30
15:00
15:30
Gbps
White Smoke Appears
2:06 pm ET
2197 Gbps Peak 3:25 pm ET
New Pope Named 3:15 pm ET
9.2 New Pope Chosen
On February 11, Pope Benedict XVI announced49
that he would
be resigning effective February 28, citing his advanced age and
lower energy levels as key factors in his decision. Upon his an-
nouncement, published reports50
indicated that the conclave to
choose the next Pope was expected to convene in mid-March,
with the intent of selecting a new Pope in time to preside over
Easter Mass. The conclave began on March 12, and was a sig-
nificant media event, both for traditional mainstream broadcast
media, as well as online outlets, with observers watching closely
for signs of white smoke, signaling the selection of a new Pope.
9.3 Syria
After seeing a number of Internet disruptions during 2012,
connectivity in the country saw problems right as 2013 began
as well. Monitoring of HTTP traffic delivered from the Akamai
Intelligent platform to users in Syria showed three brief out-
ages that occurred on January 1, between 16:00-18:00 UTC,
as seen in Figure 37. This observation is corroborated by out-
ages seen by Internet monitoring firm Renesys, which noted52
that it saw brief outages occur at 16:46 UTC, 17:15 UTC, and
17:39 UTC, with all of the impacted networks reaching the In-
ternet through the Syrian Telecommunications Establishment.
Less than a week later, Syria’s Internet connectivity saw
another disruption. As shown in Figure 38, Akamai HTTP
traffic to Syria dropped off just before 9:00 UTC on January 6,
and remained out for approximately three-and-a-half hours.
A published report notes53
that this outage occurred at the

same time that Syrian President Bashar al-Assad gave his first
televised speech since June 2012. In the days after connec-
tivity returned, however, traffic patterns showed lower daily
peaks than were seen before the disruption, indicating either
a partial restoration of connectivity, or possibly the implemen-
tation of content filters, blocking access to certain customer
content that is delivered by Akamai.
Figure 37: HTTP Traffic Served to Syria, January 1, 2013
01-Jan
00:00UTC
01-Jan
06:00UTC
01-Jan
12:00UTC
01-Jan
18:00UTC
Mbps
Finally, on January 20, another short disruption, lasting ap-
proximately half an hour, was observed. As seen in Figure 39,
Akamai HTTP traffic to Syria dropped to zero just after 21:00
UTC, returning a short time later. It is also interesting to note
that traffic levels showed a decline in the two hours ahead of
the disruption, and remained below their normal range for
approximately eight hours after the disruption as well.
Figure 38: HTTP Traffic to Syria, January 6, 2013
06-Jan
00:00UTC
06-Jan
03:00UTC
06-Jan
06:00UTC
06-Jan
09:00UTC
06-Jan
12:00UTC
06-Jan
15:00UTC
Mbps
SECTION 9:
Internet Events Disruptions (Continued)
DID YOU KNOW?
Internet monitoring firm Renesys classifies coun-
tries with only one or two Internet providers at
their “international frontier” as being at severe
risk of Internet disconnection. Syria is one of the
countries at severe risk, as are 60 other countries
including Tunisia, Turkmenistan, Libya, Ethiopia,
Uzbekistan, Myanmar, and Yemen.
[Source: http://www.renesys.com/2012/11/could-it-happen-in-
your-countr/]

9.4 SEACOM
On March 22, it was reported54
that undersea cable provider
SEACOM had experienced a number of cable cuts off the
northern coast of Africa, which affected Internet connectivity
to countries in Africa, the Middle East, and Asia. A subsequent
statement55
issued by the CEO of SEACOM noted “I know that
many of you are keen to know the cause of the outage is a
physical cable cut some kilometers north of the coast of Egypt
in the Mediterranean Sea,” and posited that “We suspect,
based on our experience with sub-sea systems and the nature
of the sea area where the cut has occurred, that the most likely
cause is external aggression to the cable most probably caused
by a larger vessel dragging its anchor across the sea bed.”
Repairs to the cable were reported56
to have been completed
during the first few days of April.
The cuts appear to have occurred just after 6:00 UTC, and
Figure 40 illustrates the impact of the cuts on HTTP traffic
levels from the Akamai Intelligent Platform to users in Ethiopia,
Mozambique, and Tanzania — three countries that leverage the
cable for international Internet connectivity.57
As shown in the
graphs, ahead of the cuts, traffic patterns in all three coun-
tries appeared to be following a daily cycle similar to the ones
seen in prior days. However, when the cuts occurred, Akamai
traffic to all three countries dropped sharply, though it did not
disappear completely. Traffic levels remained much lower than
expected in the hours following the event, at about two-thirds
of expected levels in Ethiopia and Mozambique, but just one-
third in Tanzania.
9.5 SEA-ME-WE 4 Cable Cut
On March 27, it was reported58
that the South East Asia-Middle
East-Western Europe 4 (SEA-ME-WE 4) cable was cut, which im-
pacted Internet connectivity in a number of countries that con-
nect to the cable. Published news reports59
indicated that Egypt’s
Coast Guard apprehended three divers suspected of cutting an
undersea Internet cable. The cable was also damaged in April
2010 and December 2008, both times also disrupting interna-
tional Internet connectivity for countries that rely on the cable.
Figure 41 shows daily traffic patterns for Egypt, India, and
Pakistan for several days ahead of the cable cut, as well about
a day and a half afterwards. As can be clearly seen in the traffic
graphs for all three countries, a disruption occurs at just after
6:00 UTC — published reports place the time of the failure at
6:20 UTC. It is interesting to observe the changes in traffic
patterns to these countries after the cut occurred. Of the three
countries, India appeared to be impacted the least, though
traffic levels did remain lower that day than in previous days.
In Egypt, traffic appeared to drop to near-zero levels for a few
hours after the cut. For the remainder of the day, traffic levels
peaked at approximately half the volume seen in previous days.
Pakistan also saw several significant drops in traffic around the
time of the cut, while peak traffic levels in the days following
the event were only about three-quarters as high as in previous
days. Given disruptions to this, and other, undersea cables in
the past, it would appear that Internet providers within these
countries are pursuing a strategy that includes geographic diver-
sity of the use of such cables, so that the failure of a single path
is not catastrophic.
Figure 39: HTTP Traffic Served to Syria, January 20, 2013
20-Jan
00:00UTC
20-Jan
06:00UTC
20-Jan
12:00UTC
20-Jan
18:00UTC
21-Jan
00:00UTC
Mbps

20-Mar
12:00UTC
21-Mar
00:00UTC
21-Mar
12:00UTC
22-Mar
00:00UTC
22-Mar
12:00UTC
Mbps
Ethiopia20-Mar
12:00UTC
21-Mar
00:00UTC
21-Mar
12:00UTC
22-Mar
00:00UTC
22-Mar
12:00UTC
Mbps
Mozambique
Figure 40: Impact of SEACOM Submarine Cable Outage
beginning around 06:00 GMT March 22, 2013
20-Mar
12:00UTC
21-Mar
00:00UTC
21-Mar
12:00UTC
22-Mar
00:00UTC
22-Mar
12:00UTC
Mbps
Tanzania
25-Mar
00:00UTC
26-Mar
00:00UTC
27-Mar
00:00UTC
28-Mar
00:00UTC
Mbps
Egypt
25-Mar
00:00UTC
26-Mar
00:00UTC
27-Mar
00:00UTC
28-Mar
00:00UTC
Mbps
India
Figure 41: Impact of SEA-ME-WE4 Cable Cut. Outage begins
around 06:20 GMT March 27, 2013
25-Mar
00:00UTC
26-Mar
00:00UTC
27-Mar
00:00UTC
28-Mar
00:00UTC
Mbps
Pakistan
SECTION 9:
Internet Events Disruptions (Continued)

SECTION 10:
Appendix
Unique IP
Addresses
Avg. Connection
Speed (Mbps)
Peak Connection
Speed (Mbps)
% Attack
Traffic
% Above
10 Mbps*
% Above
4 Mbps*
Region
Europe
Austria 0.1% 2,561,887 7.9 30.6 15% 78%
Belgium 0.1% 4,801,100 7.4 38.0 19% 76%
Czech Republic 0.2% 1,977,952 9.6 35.5 23% 81%
Denmark 0.1% 2,869,871 8.2 29.5 20% 77%
Finland 0.1% 2,800,533 7.7 29.2 21% 66%
France 0.6% 26,992,978 5.2 23.5 5.2% 53%
Germany 1.1% 37,840,924 6.9 30.8 13% 72%
Greece 0.1% 3,130,624 4.7 25.5 3.2% 44%
Hungary 0.8% 2,888,175 6.6 35.9 12% 75%
Iceland 0.1% 167,782 6.3 29.3 – 51%
Ireland 0.1% 1,739,818 7.3 30.9 12% 56%
Italy 1.1% 20,232,010 4.4 21.8 3.2% 35%
Luxembourg 0.1% 175,912 5.3 21.5 – 57%
Netherlands 0.3% 8,880,250 9.9 38.2 29% 84%
Norway 0.1% 3,659,326 7.4 28.7 20% 51%
Poland 0.7% 9,184,728 6.2 32.2 13% 58%
Portugal 0.1% 3,332,758 5.3 34.5 6.0% 61%
Romania 2.0% 3,019,603 7.5 47.9 19% 73%
Russia 2.7% 18,240,078 6.0 29.3 12% 63%
Slovakia 0.1% 1,033,864 6.4 29.1 9.6% 52%
Spain 0.4% 13,826,284 5.2 31.3 5.8% 55%
Sweden 0.2% 6,918,024 8.9 34.9 25% 63%
Switzerland 0.1% 3,530,481 10.1 40.3 30% 88%
Turkey 4.5% 9,654,262 3.1 25.0 1.0% 16%
United Kingdom 0.6% 28,524,028 7.9 36.3 20% 73%
Asia/Pacific
Australia 0.2% 8,635,075 4.7 26.3 4.8% 37%
China 34% 110,473,009 1.7 8.3 0.2% 5.3%
Hong Kong 1.6% 2,997,610 10.9 63.6 34% 78%
India 2.6% 16,281,136 1.3 10.6 0.3% 2.4%
Indonesia 21% 5,157,140 1.5 12.8 – 3.7%
Japan 0.9% 42,052,616 11.7 50.0 43% 79%
Malaysia 0.2% 2,252,932 2.7 23.6 1.5% 21%
New Zealand 0.1% 2,050,598 4.4 20.2 3.7% 42%
Singapore 0.1% 1,466,868 6.9 41.1 16% 62%
South Korea 1.4% 21,412,948 14.2 44.8 50% 87%
Taiwan 2.5% 12,370,121 4.3 31.3 4.5% 33%
Vietnam 0.7% 5,146,176 1.5 11.6 – 1.9%
Middle East Africa
Egypt 0.3% 3,068,473 1.1 9.3 – –
Israel 0.2% 2,501,236 7.0 35.9 12% 73%
Kuwait 0.1% 979,795 2.1 19.8 – –
Saudi Arabia 0.1% 3,855,115 2.0 12.9 – 2.7%
South Africa 0.1% 6,965,183 2.1 7.6 1.5% 8.0%
Sudan 0.1% 295,030 0.8 6.1 – –
Syria 0.1% 591,025 1.9 7.9 – –
United Arab Emirates (UAE) 0.1% 1,454,999 6.2 n/a 13% 62%
Latin South America
Argentina 0.6% 7,881,057 2.1 15.5 0.6% 10%
Brazil 2.2% 26,442,198 2.3 18.9 0.7% 14%
Chile 0.2% 3,790,528 3.0 20.4 0.9% 18%
Colombia 0.2% 5,802,203 2.8 16.0 – 13%
Mexico 0.4% 12,015,461 3.3 17.5 0.8% 20%
Peru 0.2% 1,134,543 2.0 15.2 – –
Venezuela 0.6% 2,944,646 1.1 8.2 – 1.2%
North America
Canada 0.4% 14,105,180 7.8 34.2 19% 77%
Costa Rica 0.1% 417,373 2.1 12.8 – –
United States 8.3% 147,940,918 8.6 36.6 25% 70%
*Countrieslistedwith“—“hadfewerthan25,000uniqueIPaddressesconnectingtoAkamaiduringthefirstquarteratthisspeed.Basedontherevisedthresholdforinclusion,theywerenotincluded.

SECTION 11:
Endnotes
1
https://isc.sans.edu/port.html?port=6882 (Start Date 2013-01-01, End Date 2013-04-01)
2
http://tools.ietf.org/html/bcp38
3
http://www.eassy.org/network_overview.html
4
http://www.ace-submarinecable.com/ace/default/EN/all/ace_en/index.htm
5
http://www.zdnet.com/chinas-internet-population-surges-to-564-million-75-percent-on-mobile-7000009813/
6
https://www.arin.net/knowledge/rirs.html
7
http://www.potaroo.net/tools/ipv4/
8
http://www.apnic.net/publications/research-and-insights/geoff-huston
9
https://isc.sans.edu/ipinfo.html?ip=197.228.0.0
10
11
12
13
14
15
16
17
18
https://isc.sans.edu/ipinfo.html?ip=179.176.0.0, https://isc.sans.edu/ipinfo.html?ip=179.180.0.0, https://isc.sans.edu/ipinfo.html?ip=179.184.0.0
19
20
21
https://isc.sans.edu/ipinfo.html?ip=179.200.0.0, https://isc.sans.edu/ipinfo.html?ip=179.204.0.0
22
https://isc.sans.edu/ipinfo.html?ip=179.208.0.0, https://isc.sans.edu/ipinfo.html?ip=179.212.0.0, https://isc.sans.edu/ipinfo.html?ip=179.216.0.0, 1 1
23
http://he.net/about_us.html
24
http://bgp.he.net/going-native.pdf
25
http://www.worldipv6launch.org/measurements/
26
http://www.ispreview.co.uk/index.php/2013/03/itu-secretary-general-wants-world-to-have-20mbps-broadband-by-2020.html
27
http://www.fcc.gov/document/fcc-chairman-genachowski-issues-gigabit-city-challenge
28
http://news.cnet.com/8301-1023_3-57567229-93/first-steps-taken-to-build-gigabit-network-in-north-carolina/
29
http://www.buddeblog.com.au/frompaulsdesk/the-mobile-broadband-boom-in-latin-america/
30
Ibid.
31
Ibid.
32
http://www.ntia.doc.gov/blog/2013/household-broadband-adoption-climbs-724-percent
33
http://www.renesys.com/2013/01/cuban-fiber-completo/
34
http://www.telegeography.com/products/commsupdate/articles/2013/01/25/cuba-confirms-first-submarine-cable-is-active/
35
http://techcrunch.com/2013/03/21/china-broadband-laggin/
36
http://europa.eu/rapid/press-release_SPEECH-13-263_en.htm
37
http://www.gigaclear.com/oxfordshire-businesses-embrace-ultrafast-broadband/
38
http://www.ispreview.co.uk/index.php/2013/04/b4rn-extend-1gbps-ftth-broadband-in-lancashire-uk-to-21-parishes.html
39
http://www.theaustralian.com.au/australian-it/israel-sets-its-sites-on-its-own-national-broadband-network/story-e6frgakx-1226254885060
40
http://www.akamai.com/html/io/io_dataset.html#stat=mobile_browsertop=5type=linestart=20130101end=20130331net=mhide=ope
ra%20mini+chrome%20mobile+docomo
41
http://www.akamai.com/html/io/io_dataset.html#stat=mobile_browsertop=5type=piestart=20130101end=20130215net=m
42
http://www.akamai.com/html/io/io_dataset.html#stat=mobile_browsertop=5type=piestart=20130216end=20130331net=m
43
http://www.akamai.com/html/io/io_dataset.html#stat=mobile_browsertop=5type=linestart=20130101end=20130331net=bothhide=chro
me%20mobile+diigo%20browser+opera%20mini
44
http://www.akamai.com/html/io/io_dataset.html#stat=mobile_browsertop=5type=piestart=20130101end=20130215net=both
45
46
47
48
http://www.bbc.co.uk/news/world-latin-america-21679053
49
http://articles.washingtonpost.com/2013-02-11/national/37027069_1_georg-ratzinger-paolo-gabriele-pope-benedict-xvi
50
Ibid.
51
http://en.wikipedia.org/wiki/Papal_conclave,_2013
52
http://b2b.renesys.com/eventsbulletin/2013/01/SY-1357062270.html#more
53
http://thinkprogress.org/security/2013/01/06/1403061/syria-internet-outage-assad-address/?mobile=nc
54
http://www.itnewsafrica.com/2013/03/seacom-suffers-internet-outage-off-african-coast/
55
http://www.itnewsafrica.com/2013/03/seacom-outage-could-last-till-april/
56
http://www.communicationsafrica.com/internet/africa-s-broadband-returning-to-normal-following-seacom-cable-repair
57
http://www.businessdailyafrica.com/Corporate-News/Seacom-cable-cuts-disrupt-telcos-internet-links/-/539550/1727582/-/5m78ckz/-/index.html
58
http://gigaom.com/2013/03/27/undersea-cable-cut-near-egypt-slows-down-internet-in-africa-middle-east-south-asia/
59
http://www.reuters.com/article/2013/03/27/net-us-egypt-internet-idUSBRE92Q1AQ20130327

@Akamai
facebook.com/AkamaiTechnologies
linkedin.com/company/Akamai-Technologies
youtube.com/AkamaiTechnologies
blogs.akamai.com
pinterest.com/AkamaiTech
slideshare.net/AkamaiTechnologies
WE’RE SOCIAL, FOLLOW US!

©2013 Akamai Technologies, Inc. All Rights Reserved. Reproduction in whole or in part in any form or medium without express written permission is prohibited. Akamai and the Akamai wave logo are registered
trademarks. Other trademarks contained herein are the property of their respective owners. Akamai believes that the information in this publication is accurate as of its publication date; such information is subject
to change without notice.
Asia Pacific Headquarters 1 Raffles Place, #16 – 61 One Raffles Place, Singapore 048616 Tel +65.6593.8700 Fax +65.6593.8799
European Headquarters Pfingstweidstrasse 60, 8005, Zurich, Switzerland Tel +41.43.210.91.00 Fax +41.43.210.91.01
North American Headquarters 8 Cambridge Center, Cambridge, Massachusetts, United States 02142 Tel +1.617.444.3000 Fax +1.617.444.3001
Akamai®
is the leading cloud platform for helping enterprises provide secure, high-performing user experiences on any device, anywhere. At the core of the company’s solutions
is the Akamai Intelligent Platform™ providing extensive reach, coupled with unmatched reliability, security, visibility and expertise. Akamai removes the complexities of connecting
the increasingly mobile world, supporting 24/7 consumer demand, and enabling enterprises to securely leverage the cloud. To learn more about how Akamai is accelerating the
pace of innovation in a hyperconnected world, please visit www.akamai.com and follow @Akamai on Twitter.
Acknowledgements
EDITOR: David Belson CONTRIBUTOR: Michael Smith CONTRIBUTOR: Mathias Sintorn (Ericsson)
ART DIRECTOR: Brendan O’Hara CONTRIBUTOR: Martin McKeay CONTRIBUTOR: Geoff Huston (APNIC)
GRAPHIC DESIGNER: Jason Thornton CONTRIBUTOR: Martin Levy (Hurricane Electric)
CONTRIBUTOR: Jon Thompson CONTRIBUTOR: Svante Bergqvist (Ericsson)
Please send comments, questions, and corrections to stateoftheinternet@akamai.com
Follow @akamai and @akamai_soti on

The State of Internet

More Related Content

What's hot (17)

Viewers also liked (8)

Similar to The State of Internet (20)

More from Dijital Politikalar Akademisi (9)

Recently uploaded (20)

The State of Internet