The edge isn’t the only technology play looking to master latency

• Elon Musk views low latency as the key Starlink satellite attribute 
• Sending signals through space is way faster than pushing them through fibre
• Will Starlink play a big part in enabling next generation low latency transmission?

Latency (or rather its lack) is apparently the key business driver for the multi-billion dollar investments in the future mega orbital constellation SpaceX Starlink, brainchild of ‘Space-Karen’ Elon Musk. 

SpaceX Starlink is a fantastically ambitious play. Eventually 12,000 satellites will be operating and will be capable of covering just about the entire globe  for broadband transmission. (see this TelecomTV article from 2018 on LEOs and satellite history). 

Starlink connections won’t be impossibly expensive either - even at  this early stage in the ‘beta’ phase, service is available for $99 a month.

There’s been plenty of noise about SpaceX and Starlink, but the noteworthy thing right now is this emphasis on latency rather than just the broadband capability. 

Musk reckons all sorts of wonderful new applications come into view when you can deliver data for a handful of milliseconds. To a TelecomTV reader/view, that might sound like an extremely familiar observation - it’s been the clarion call for 5G for at least 5 years.  But Musk is talking about very long distances between, say, one side of the Atlantic to the other, where stock price information exchanged quickly (for instance) is worth millions .But there are many other applications which look like they’re in need of a low latency pick-me-up too. It could be that Starlink or other LEOs, will play a valuable role in long distance latency-crushing alongside edge facilities and MEC (multi-protocol edge computing). 

Satellite capabilities seem to have come full circle. In the early 1980s the dominant long distance telephone call was made by satellite and a very frustrating experience it was too, with lots of long pauses and clashing voices. That changed with submarine fibre when long distance conversations became practically normal-sounding. Now satellites are back with Musk touting their low latency credentials. 

The difference is orbital height of course. Musk reckons space transmission from LEOs will simply have the latency ‘edge’ (as it were) over undersea fibre, since light goes faster in a vacuum, even with relatively high orbiting LEOs. Meanwhile, no matter how clean or advanced several thousand miles of undersea fibre can be made, they won’t be as clean as space. A compelling video explainer on Starlink is worth watching here. 

Starlink already exists, but with nothing like the 12,000 satellites that Musk currently envisions. However, Musk is the master of the hype driven beta period and damn the torpedoes. So customers using Starlink's beta service pay $99 per month and $499 for a starter kit that includes the full phased-array antenna which tear-down experts believe probably costs SpaceX thousands of dollars a piece. 

But this is the IT industry where hardware costs can always be buried by volume... eventually. So the cost of the first few thousand customers will be  worth it if it means having happy customers singing the system’s praises (mostly) before there’s even a full launch. 

What about the space junk problem? There’s a plan for that and it seems to have convinced regulators. If Elon can’t deliver you’d better believe they’ll pull the plug and his billions will be down the drain - many of us are inclined to believe it will work or at least has a very good chance of working. He may be the ‘Space Karen’ but when he calls the manager he usually gets his way.