Tag Archives: liquid telecom

Ten Years Of Supporting Africa’s Tech Community

In 2010, Liquid Telecom was approached to support the first Africa Peering and Interconnection Forum (AfPIF). It was a great opportunity to start conversations on how to share content locally and offer better user experience. We’ve continued to support Africa’s tech ecosystem ever since – the Liquid Telecom team has attended every Afpif event and we’ve been a platinum sponsor since 2012.

One of the big issues back in 2010 was the amount of content that was hosted abroad – meaning higher transit costs, which can lead to higher connectivity costs to the end user. The goal was to increase the level of content exchanged locally to 80%. This meant we had to invest in more data centres and cloud providers, as well as educate the users on the importance of hosting locally.

Experts at the early AfPIF events predicted a more interconnected Africa and that by 2020, 80% of all Internet traffic downloaded in Africa will be sourced from a data centre or exchange point in Africa.

For Liquid Telecom that advice has come true and we achieved the milestone in 2019, but it has not happened for all countries in Africa which we are yet to reach. 

Four or five larger African owned multi country backhaul networks have emerged, all peering with each other at strategic hub locations to exchange traffic locally and keep African data in Africa. At Afpif meetings we learnt as an industry that if we peered locally with each other that we would flourish faster than those that staunchly refused to peer. 

We also learnt that by doing this, our costs of transporting data to and from Europe would reduce significantly, and even that content giants would begin to pay to bring their content to Africa – instead of Africa paying to get to them. The entry of the global Internet giants into Africa – who have decentralised the infrastructure of their platforms and moved from centralised data centres to a distributed model of multiple data centres and caching nodes connected by robust transport networks – will play a big role in achieving that 80% figure.

But what is next for Africa’s Internet?

Certainly, the African Internet ecosystem looks set to enlarge as more countries take progressive steps to open up their telecoms market and improve connectivity.  Liquid Telecom just announced the construction of fibre to Juba in South Sudan, reducing the small number of countries that still have no backbone fibre optic network to their capital cities.  And there are positive signs that the liberalisation of telecoms in Ethiopia, a country of over 100 million people, looks to be happening right now.

The Africa free trade agreement has been ratified and finally signed by Nigeria leaving only one country, Eritrea, being not part of the world’s largest trading area. African Union predicts that intra African trading will increase from current 16% by a factor of 60% by 2022. Intra African trade sets to drive more investment in road, rail and ICT infrastructure to facilitate the growth, and the usage on the digital infrastructure sets to increase, further stimulating the trade by selling of digital assets and services across borders.

But it’s not all good news. We have seen backward-looking regulations passing in some countries to limit or regulate popular Internet applications such as social media. Internet shutdowns in Africa have become more prevalent and serve to harm the economies of those countries involved. With more digital transformation and cross border interconnection, it maybe only a matter of time before the impact of a national Internet shutdown will affect users of digital services in neighboring countries. 

Global politics also comes into play and a trade war between US and China, now being aimed at a vendor who has been successful in Africa and instrumental in helping governments and MNOs roll out African fibre and 3G/4G services, can’t be good for Africa either.

But overall, we expect to see strong growth in additional infrastructure and more interconnection. This means more access, more applications, more cloud being consumed and more digitisation of business and government services, leading to more completely new digital services being consumed by a rising African youth population. Not to mention the potential explosion of connected IoT devices.

By 2030, there will be multiple ways for Internet traffic to travel from Cape to Cairo within Africa, and multiple fibre routes across the region. All countries in Africa will be connected to fibre and interconnected, and much investment will have been made in last mile access. Many countries will have carrier neutral data centres, but it is likely that the biggest concentration of data centres and cloud computing will be in those countries that have larger digital economies and safe regulatory environments for data storage. Interconnection will enable the smaller countries to access Africa’s Cloud.

From a technology perspective some key developments are happening behind-the-scenes. Afrinic is the last regional Internet registry to have remaining IPV4 addresses and we are now looking at a nearly empty store. The African Internet of the future will therefore be advancing rapidly towards IPV6. 

There have also been some key developments in IP packet routing and segment routing that are enabling us to route traffic based on where it has come from as well as where it is going to. We at Liquid Telecom were one of the first globally to deploy segment routing as well as one of the first in Africa to deploy IPV6 to the last mile customer endpoint – and it was great to receive an award recognising that in London at the end of 2018. We are also looking to the future and getting involved with standards development, especially where segment routing and IPv6 converge, in order to ensure the standards of tomorrow cater for the needs of Africa and its customers.

These technology enhancements will pass on benefits to our customers. Increasingly our larger backbone users are wanting to migrate towards IP based services from traditional IPLC leased lines, but still want control and flexibility of the exact routing and path taken. By 2030, not only will there be multiple terrestrial paths for Internet traffic to travel from Cape to Cairo or from east to west, they will also likely be carrying transit traffic direct from Asia to the Americas. Network providers will adopt new routing technologies to deliver more choice to their customers about where their traffic comes from and how it reaches its destination across the multitude of paths available on the continent – so that they can mitigate their risks and enjoy an Internet experience tailored to their requirements.


Image credit: Internet Society / Nyani Quarmyne / Panos Pictures

Demand for World Cup Streaming Exposes lack of Local Content

If there is one thing we picked from the 2018 World Cup in Russia is that video streaming of entertainment content in developing countries is approaching mainstream.

As Internet penetration continues to skyrocket and the price of Internet enabled devices and data reduces, more people in developing countries today are live-streaming their entertainment demands.

This has been evidenced by the growing popularity of video streaming services like Netflix, YouTube and Showmax among young people in developing countries.

These platforms appeal especially to young people because of their vast repositories that offer content from across the globe at a few keystrokes. Netflix users alone collectively watch over 1 billion hours of content each week while it would take over 60,000 years to watch everything on YouTube.

These stats were demonstrated during the just concluded World Cup where millions of people across the world plugged in to catch up with the exhilarating action from 64 matches across stadiums in Russia.

Analysis by Internet traffic monitoring site Akamai indicate that it took just ten days to surpass the previous video streaming record set at the 2014 World Cup. By the end of the group stage, Akamai had streamed 65% more data from start to finish than it did in Brazil four years earlier.

Akamai reported the highest number of concurrent streams peaking at 9.7 million during the Mexico v/s Sweden match at the same time as South Korea lined up against Germany on 27 June.

This, compared to the 5 million viewing peak for the entire 2014 World Cup matches indicates an unprecedented rise in the number of online viewers from developing countries. The peak bandwidth for streaming in the first round in Russia was 23.8 Tbps, compared with 6.99 Tbps during Brazil’s World Cup.

Kwese and Kwese iflix, one of the continent’s fastest growing streaming services, has developed video streaming partnerships with mobile network operators across Africa (during the world cup season) and according to Mr. Ben Roberts, CTO of Liquid Telecom, the number of number of video streaming customers has risen as a result.

Telecommunications service provider Liquid Group provides host infrastructure for Kwese and Kwese iflix and Mr. Roberts noted that “Streaming is steadily growing, getting more and more each day”.  “The most popular was the Nigeria vs Argentina (not surprising), but also Germany and Brazil’s last game of the first stage exceeded previous games.”

However Mr. Roberts states streaming among African users has been heavier on matters that touch closer home.

“It’s not the biggest ‘internet event’ in Africa we have seen this year, with the traffic around the coup in Zimbabwe and subsequent resignation of Robert Mugabe being something that turned up the traffic on all channels and links across Africa to a very noticeable degree,” he said.

This means while African consumers demand and are ready to spend on video streaming, the lack of local content is a gap that provides opportunities for developers and creators.

More than 90 percent of African Internet content is hosted outside the continent and this means server request from the continent take much longer because they have to go through exchange points in Europe or North America. Currently more than 57 percent of Kenya’s web content is hosted in North America and 25 per cent in Europe with just 10 per cent in Africa according to data from Alexa.

This implies a longer turnaround for connectivity requests at the same time and slower download speeds.

Some streaming service providers have tried to create local hosting sites. Early last year ShowMax signed a partnership with SEACOM to have its servers hosted in Nairobi to better serve its East African audiences.

This is however only a stopgap measure because the content available to users is still largely North American and European. African content creators need to rise up to the challenge of providing the entertainment needs of a hungry, tech savvy and discerning population.

Read or download our report –  Promoting Content In Africa

Why more is better when it comes to subsea cables and Africa

By Ben Roberts, Liquid Telecom

Between 2009 and 2012, seven major subsea cables were deployed along the east and west coasts of Africa, bringing an abundance of international connectivity to the continent for the first time.

An estimated $3 billion poured into the construction of these undersea networks, which have played an important role in developing Africa’s internet ecosystem.

Today, most of these cables are less than 7 years old – to add some perspective, the average lifespan of a subsea cable is 25 years – and not all of their capacity has been lit.

Meanwhile, a new generation of subsea cables are making their way to Africa.

Earlier in the year, Liquid Telecom announced its first subsea cable project, called Liquid Sea, which will run the length of Africa’s east coast with onwards connectivity to Europe.

We are not alone in this endeavor – other projects in the pipeline include the Djibouti-Africa Regional Express (DARE), as well as the consortium backed Africa-1 and O2Cs. There is even a proposed project to connect Africa to Latin America with a subsea cable link for the first time; the South Africa Cable System (SACS).

This has prompted cynicism from some industry leaders and market watchers, who suggest that all these new systems are unnecessary.

Africa’s phenomenal demand for internet access suggests otherwise – and efforts are currently underway to ensure that 80% of all African internet content is being served from Africa.

Here are six reasons why more subsea cables are beneficial to the market and the continent as a whole:

1)         Connecting the unconnected: Not all African countries were connected by the first wave of subsea cables. Some countries such as Eritrea and Somaliland were overlooked, while there is fresh demand for access to cables from other nations – for example, northern Mozambique requires reliable high-speed internet to support the rise of new oil and gas reserves.

2)         Greater diversity: At the moment, there are a limited number of African landing stations where the major subsea cable systems interconnect, and traffic is sent onwards to Europe and Asia. This enables certain companies to act as ‘gatekeepers’ and charge excessive cross-connect fees, which in some cases can be about 50% of the total cost of bandwidth sold in Africa – just for provisioning a 30 metre piece of fibre optic cable to a landing station. New cables bring diversity and increase choice so that traffic can be switched to the best location, as well as avoid excessive fees.

3)         Increasing competition: More subsea cables and landing stations will bring greater competition and geographic redundancy. Countries such as the Seychelles are currently reliant on one single subsea cable, with satellite as the only back-up. Elsewhere, countries reliant on one landing station owned by one company will always suffer from high bandwidth pricing. Two landing stations will help improve resiliency and pricing, but four or five landing stations will really start to drive pricing down as well as improve reliability.

4)         New players in the market: There’s been dramatic changes in the market since the South Atlantic-3 was built in the early 2000s by a consortium of copper network and fixed-line operators. Today’s telecoms ecosystem is much more diverse and features many more players (including OTTs), which were unrepresented in the first wave of subsea cable systems. These new players require capacity and are investing in new systems accordingly.

5)         Advancements in technology: Technology is advancing all the time, but many of the older cable systems operate using SDH-based networks, with repeater spacing and fibre types that prevent upgrades. More modern systems can upgrade using 100G wavelengths, but new systems can leverage this mega capacity technology from day one.

6)         Keeping engineers in a job: The world has to keep building new cables in order to keep grey-haired engineers employed. Actually, they should hold off retirement just yet as the subsea cable industry is enjoying a major renaissance at the moment. We’ve seen lots of investment worldwide in new projects, and competition is heating up between vendors. In fact, many of them are offering great prices and many projects are also receiving government funding. There’s perhaps never been a better time to be a consultant or an engineer.