The Ripe Atlas Internet measurements infrastructure in Africa is about to receive a significant boost, thanks to the arrival of a brand-new Anchor in Burkina Faso! The Anchor, to be hosted by Orange Burkina Faso S.A., will now put the number of Ripe Atlas Anchors in Africa at eight (8). At present, one Anchor is in North Africa in Tunisia, three in East Africa (Kenya, Tanzania and Mauritius), and three in South Africa. It is exciting that, with the coming live of the Anchor in Burkina Faso, there will now be at least one Anchor in the West Africa region. Many thanks to Jean Baptiste Millogo and his Orange team, for their tremendous efforts to get the Anchor into Burkina Faso! Map below shows the global geographical distribution of Anchors.
Geolocation of Ripe Atlas Anchors, 227 in total
But why is it necessary to have Ripe Atlas Anchors in all the regions? For starters, apart from functioning as enhanced vantage points with more measurement capacity than regular Ripe Atlas probes, more importantly, the Anchors act as regional targets for measurements performed by hundreds of probes from the Atlas network, thereby providing valuable information about local and regional connectivity for the entire Internet community. What this means is regional Internet performance data is only collected for those regions with Anchors, while Internet performance challenges and disruptions in the regions that lack Atlas Anchors may go unnoticed. Anchors act as well-known and cooperating targets for user-defined measurements, and are capable of handling more extensive measurements, as they are located within networks that have sufficient bandwidth to support a larger number of incoming measurements. Just like regular Atlas probes, Atlas Anchors can also be used as sources and targets for ping, traceroute, DNS, HTTP(S) measurements.
Anchors are particularly valuable for researchers in that Anchor names are encoded in such a way that they conveniently provide geolocation information, including their country and city location, as well as the Autonomous System Number in which they are hosted. This provide Atlas users with the ability to choose vantage points or regional targets in different networks, cities and countries. Furthermore, while regular Atlas probes are often offline and therefore less reliable for scheduled experiments, Anchors tend to be more reliable as they are online most of the time, considering that they are hosted in server environments with high availability rates.
The advantage of hosting an Anchor is that, as a host, you get access to the results of Anchoring measurements (ping, ping6, traceroute, traceroute6, HTTP) towards your Anchor from all the other Anchors and several hundred probes, providing you a continual overview of your regional reachability. The Anchor hosts also get to accumulate plenty of Atlas credits, which they can use to execute their own custom measurements on the Atlas platform. What more, the hosting organizations get promoted by having their logos displayed on the Atlas Community page.
While this addition is very good news, it is far from being enough. For context, globally, there are 277 Atlas Anchors, and only seven of these are in Africa. We need more Anchors in Africa so that better and more accurate continental Internet performance data can be obtained. Ideally, we would need at least one Anchor in each and every Africa country. It will be great to have more networks hosting Atlas Anchors, especially in African countries without Anchors at present. Network requirements for hosting an Anchor include the need for native IPv4 and IPv6 (if IPv6 is announced in the host ASN); static IPv4 and IPv6 addresses need to be unfiltered (not firewalled); a single RJ45 network connection; and the ability to provide up to 10 Mbit bandwidth. More details about becoming an Anchor host can be found here.
