The Latency Conundrum: Setting the record straight

05 Aug 2019

Speed testing is commonplace in South Africa, with several online tools available to measure Internet speed in addition to regular well-read reports from local publications. This fascination with connectivity speed probably stems from our history of bandwidth-limited legacy technologies coupled with highly oversubscribed networks, which have led to a great deal of frustration and distrust on the part of consumers in the past.

Today, the same interest in line speed is reinforced by the competitiveness of the ISP marketplace – especially in regard to both Fibre-to-the-Home (FTTH) and Fibre-to-the-Business (FTTB) package options. In an environment where providers are continually trying to out-market each other, and bandwidth accessibility is exploding, people are demanding the lightning-fast speeds promised by providers. Speed tests and their findings appear then as empirical proof to help customers ensure their rightful piece of the high-speed connectivity pie.

Except, speed tests are too often flawed when it comes to accurately measuring ISP performance. Here’s why:

It may be convenient to consult a table neatly outlining the average upload and download speeds, as well as latency, on fibre and mobile lines from the nation’s biggest providers. The issue, though, is that these speed tests may not be adjusted to factor in all the complexities of performance assessment. This is especially true in the context of South Africa, located as it is on the southern tip of the African continent.

Since 2016, South African companies have been shifting their storage, security, communications and other business tools from on-premise to the cloud. Microsoft and Amazon may have arrived on our shores, but their data centres are largely serving a network function, for now, as opposed to hosting cloud services themselves. The point is that the cloud – and the servers that host the cloud’s functionality – are still physically far away in Europe and North America. Unsurprisingly, this long-distance digital migration has been accompanied by more complaints about connectivity speed as content must still be largely accessed from thousands of kilometres away.

Long-distance links, like those between Cape Town and London, for example, are high capacity by nature. Because of the distance they cover, they are also high latency – where the time it takes data to travel from one end of the link to the other, and back again, is greater. For the record, latency, also known as Road Trip Time (RTT) or delay, is generally regarded as becoming noticeable when it crosses the 80 milliseconds threshold.

When a conventional speed test is performed, the device is generally not set up to account for the innate latency of the system, nor how to overcome it. This optimised setup is not default, however. It requires a relook of system settings to achieve it. For South African customers to enjoy the most satisfying connectivity experience then, it’s time to stop focusing on line speed to the exclusion of everything else – especially when measures are not necessarily an accurate reflection of reality. By all means, use speed as a factor to guide your decision, and raise concerns with providers about improving line performance. But other factors, like service uptime, application performance and ISP support, are just as, if not more, important to ensure a solution that meets all of your requirements, whether for business or home.