For many developing countries, a lack of legacy architecture has proven to be a bonus rather than a disadvantage in terms of closing the digital gap. As the digital economy accelerates across much of Africa, investment is ploughed into fibre and mobile – but data centres can be overlooked, according to First Brick Holdings.
Set up as a portfolio company of US-based Roha, First Brick Holdings is focused on investment and development for data centres across Africa, although it is primarily active in the east. We spoke to the company’s executive director Robert Mullins about the dearth of data centres in much of Africa, and how this needs to be addressed.
Why have data centres been overlooked in Africa?
If you look at what’s been happening across Africa, plenty of backbone fibre and urban fibre has been put in, so from a connectivity perspective data is being transported fairly well. The last six to eight years have also seen substantial improvements in the international gateways.
There’s also been a tremendous amount of investment in the mobile infrastructure – there’s great connectivity in much of Africa, with things like video calling completely commonplace. Data is being transported effectively and efficiently, both within the borders of most African countries and outside of them. There’s much better latency than there has been in the past,
The one thing that from our perspective is being left out is what happens to the data that’s being transported in-country – that’s what we consider the third leg of the stool. This critical piece of infrastructure, processing and storing data within the borders of these countries, has been left out of the equation. Effectively, the offering of a data centre is to provide an environment in which customers from across both the public and private sector can host and keep their data and core systems working in an optimal environment, as is being done in the rest of the world outside of Africa.
Why has adoption of data centres lagged, and how has the issue been overcome?
It’s a question of priority – the competition in terms of legacy is generic. People have built their own ‘data centres’ in-house; six or eight racks in a server room for banks, or smaller for private enterprises. Telecom companies built their own for their own purposes and offered space to third parties, so they exist in a generic form but not in the way that they exist outside the region, where they’re a shared infrastructure that benefits all stakeholders.
You have a much better service and offering, as well as a lower cost of ownership, plus the additional oft-overlooked benefit: it’s an actual location with proper security and cooling, backup power etc, and the equipment is operating optimally. One of the fundamental benefits of this is that they create an environment where different customers and networks can connect to each other in more effective ways – it’s a healthier environment.
How are you encouraging operators in Africa to move towards this data centre model?
We’ve positioned it as a carrier-neutral and cloud-neutral environment, which means that all telco carriers – whether international, regional or local ISPs – can come in with their own equipment and offer choice to other customers hosting within the data centre. Neutrality is the optimal model – different cloud providers can come in and have equipment within the data centre that lets them offer services. Beyond the traditional functions of the data centre, it creates an environment of interconnectivity and choice.
We believe that this is an advantageous proposition in terms of digital transformation. When you look at what’s happening in many of these markets – for example Uganda – the nature of entrepreneurship and the way the business environment is evolving, it’s very much in a digital sense. We can create an enabling cornerstone which is hugely beneficial to the broader economy and society – not just in the context of Uganda, but across the region.
What are the key growth areas in Africa’s digital economy?
“Digital Economy” is a broad term – the economy itself is increasingly becoming digital one way or another. There will still be traditional industries such as manufacturing, and countries such as Uganda are pushing towards local manufacture, but a lot of these industries are themselves digitally transforming. This is for their own benefit in terms of internal efficiencies, but also in how they can offer access to their customers - whether this is within their countries or regionally. It gives them a chance to compete regionally and/or internationally; if there’s an infrastructure that allows them to have digital platforms for purchasing or creating a market place, that’s by extension driving growth of the traditional economy as well. It’s a sequential migration towards a broader digitalisation.
In another sense, there is a growing digital economy in its own right; a lot of the digital entrepreneurship within Uganda is focused on areas such as cloud service offerings, digital marketing, and mobile payments, so there’s a very vibrant digital economy developing.
Is it fair to say that the relative novelty of these services is a factor driving uptake in Africa?
If you look at the demographics in these markets in terms of the youth population and technology that they adopt, it’s a tremendous opportunity and in that sense it’s not that different from what we see in more developed markets – for example, I wasn’t expecting Uber to be a pretty common way to get around Kampala.
The mobile payments market is far more developed than in many more developed economies. There’s a full understanding and perception of the availability of these services because they are there, but there’s a difference between a service being present in a market and having actual availability – and the latter is often much lower than it should be due to missing infrastructure. For example, if I arrive at the airport in Kampala there are around five ATMs, but on any given day only one or two will be working, and it can be the same for mobile payments services. They’re lacking the infrastructure to the support all the services that are being demanded by the population.
Are these services comparable? Would the availability of a digital service not be more reliable?
Many of the legacy mobile payments services have been so successful in part due to using incredibly reliable technology standards such as USSD or SMS. That being said, the whole mobile payments segment is moving towards omni-channel, so while mobile banking in Africa has typically been supported by more legacy technologies, this shift towards omni-channel strengthens the case for infrastructure. The systems that are supporting ATMs are back office systems in banks, and if there are issues with how those are being run then customers can’t withdraw money. The legacy mobile payment market will be around in Africa for some time but the migration is underway.
These services are launching in more densely populated areas – is it viable to extend them to more remote regions?
From a data centre perspective you don’t need to build out into remote areas; connectivity within these countries is generally very good, or at least not a huge issue in most places. What the data centres are doing is providing a centralised infrastructure where a bank or mobile payment provider’s core systems are guaranteed to be always on. The advantage of the data centre is that due to the neutrality aspect, any of these banks or mobile payment providers can have a choice of carrier. Typically what they would do is subscribe to two or more carriers, so if there’s any issue with their connectivity, i.e. carrying that data in and out of their core systems, they have a backup channel on which they’re able to do that. In the current environment that’s typically not the case; banks usually have one service provider delivering them connectivity from their headquarters or wherever their core systems are based. This is also addressing uptime and risk from a connectivity perspective, as well as keeping the system itself up and running.
The content side of things is also driving both traffic and the need for a local cache. Activity isn’t just based around financial transactions – a lot of it is content, and there’s a tremendous amount being generated in Africa. There needs to be a place for this to be effectively stored and accessed, but additionally access to global content is also growing exponentially. From a user experience perspective, this content needs to be closer to the audience accessing it – lower latency will drive a better user experience. The thirst for content is only growing so content providers are realising the importance of having this infrastructure in-country, closer to the population.