Developing IXPs in Africa: Why it’s important and how it offers potential for growth

Developing IXPs in Africa: Why it’s important and how it offers potential for growth

Second only to Asia in landmass and population size, Africa is currently home to 1.216 billion people. That massive population is also young as 40 percent is less than 14 years old. UNICEF predicts that by 2050, 25 percent of all the people in the world (and 40 percent of the world’s youth) will be African . These young people are, like Millennials in Europe or the US, the online generation,which has helped increase the number of subscriptions to mobile services in Africa over the last few years. According to the GSMA[2], 46 percent of the African population subscribed to mobile services at the end of 2015 and it is predicted that an additional 168 million people across Africa will be connected by mobile services by 2020, taking the total to around 725 million unique subscribers.

It sounds like a connectivity success story but a reality-check from Internet World Stats 2017 shows that only 27 percent (circa 346 million people) of Africa’s population used the Internet in March 2017 – significantly less than the 54 percent internet penetration figure for the rest of the world. The real picture is of a continent with potentially huge opportunities for carriers, OTTs, Content Delivery Networks and Internet Service Providers (ISPs) and so it is no surprise that wholesale carriers serving Africa are anticipating new customers as these players explore how to maximise new market opportunities in Africa.

Challenges to meeting capacity demand in Africa

The African market is fragmented with three dominant countries – South Africa, Egypt and Nigeria – accounting for one third of the continent’s mobile Internet subscriber-base. With geography, politics and economics potentially challenging the scaling of infrastructure and implementing new services whilst delivering on performance SLAs, especially in remote locations, the question has been asked: “Is mobile the answer and might Africa become a ‘mobile first’ continent?”. It has certainly been demonstrated that developing countries, which have not followed a traditional infrastructure-building path, can ‘leapfrog’ earlier technology stages such as landlines or bricks and mortar branch banking and go direct to mobile telephony, Internet broadcasting and mobile money – but, and it’s an important caveat, only if they have the bandwidth capacity to do so.

There is unquestionably an increase in capacity demand and it is being driven in part by global OTT players and content providers including Google, Microsoft, Facebook and Netflix. Traditionally, these companies have used other organization’s networks to run their own data-intensive applications. However, by locating their content closer to local Africa markets they can improve the performance of their services and user experience and therefore increase the adoption of their services. In general, players follow a two-pronged network deployment strategy: localizing their content directly into the local markets within Africa when feasible and/or go to gateway locations, such as Marseille. The gateway model is a good one, offering reasonable network latency and a perfect network environment that combines availability and affordability of both network supply to reach all continents and international datacentres to host the content, in a fully-deregulated market.

Keeping it local – the role of the IXP in Pan-African connectivity

The Internet Society explains the role of the IXP thus : “to send a package from London to Paris and to route it via San Francisco would be foolish, unnecessarily expensive and inefficient”. Yet this is what is being done every day with Internet traffic between local Internet Service Providers (ISPs), destined for local customers. In Africa, the absence of connectivity between ISPs often results local traffic being routed over expensive international links simply to reach destinations within the country of origin. These links must be paid for in foreign currency so in effect, ISPs are paying international ‘shipping’ rates for a local delivery.

The good news is that there is an internationally recognized solution to this inefficiency, which is an Internet Exchange Point, or IXP and there are currently 37 active IXPs located in 34 cities in 28 African countries . As illustrated above, the primary role of an IXP is to avoid expensive international transit costs by keeping local Internet traffic within local infrastructure. The costs associated with traffic exchange between ISPs are reduced, because IXPs allow for the free exchange, or peering, of domestic Internet traffic. Furthermore, IXPs improve the quality of local Internet services by reducing the latency associated with unnecessary traffic routing and they can also serve as a convenient hub for hosting value-added and critical infrastructure within a country. Overall, IXPs are a crucial part of building a sustainable domestic Internet ecosystem/mobile connectivity and their creation is a positive marker of economic and social improvement.

Keeping it local and relevant

From a purely technical point of view, building an IXP is not too difficult. However, building the required level of trust and collaboration between the stakeholders is the challenge and being able to speak the same language is a good starting point. Along with English and Arabic, French is one of the principle languages in Africa with 54.7 percent of the population speaking French compared to 36.4 percent speaking it in Europe . 13 African countries have French as first official language and 16 have it as their “co-official” language . In these countries it should go without saying that it is important, from a commercial, educational and development point of view, that French language content is delivered to French speaking populations. African IXPs operating in countries with a high to medium usage of the French language currently include CASIX in Morocco, SENIX in Senegal, DjIX in Djibouti, BENINIX in Benin, KINIX in the Democratic Republic of Congo, CIVIX in Ivory Coast, TunIXP in Tunisia, BFIX in Burkina Faso, GAB-IX in Gabon, RINIX in Rwanda and MIXP in Mauritius.

Mobile can make a difference, but Internet penetration must grow

Although the Internet in Africa is growing fast there are major disparities: mobile broadband access accounts for more than 90 percent of Internet subscriptions, but the majority of countries have an Internet penetration below ten percent (lower than the 20 percent found to be critical for countries to reap the economic benefits) and users in Africa pay up to 30 or 40 times more for Internet access than their peers in developed countries.

Recognising the importance of IXPs in facilitating Internet-based economic growth, The Internet Society has an ambitious continent-wide programme for interconnection and traffic exchange (ITE) for Africa, with a goal to support 80 percent of local and 20 percent of international Internet traffic by 2020.

Many people and organizations are working to help make this happen as they understand and believe in the value of peering and interconnection in a domestic or at least, gateway, IXP. Out of the 37 African IXPs that are currently active, 21 were established within the past ten years, often with thanks to individuals and organizations who contributed to providing the equipment, training, valuable experience and more.

Initially published in Connect-World Magazine, Africa and Middle East edition, on October 10, 2017.

Leave a Reply

Your email address will not be published. Required fields are marked *

+ 87 = 88

massa quis, tempus ipsum dictum consequat. elementum felis commodo id, Nullam mi,