Ask HN: Are you prepared for loss of multiple trans-atlantic fibre-optic cables?
Is this something we as an industry should be working together on to understand the risk and how we may be able to cooperate with each other to mitigate the risks and recover rapidly (such as pre-positioning cable repair ships at times of elevated risk) ?
As the situation in Ukraine vis-a-vis an invasion by Russia deteriorates [0] Russia has today told the Irish government it intends to conduct naval live-fire exercises February 3-8 in a small area of the Irish Exclusive Economic Zone (EEZ) 160 nautical miles (~240km) SSW, off Mizen Head [1], coincidentally where the major fibre optic cables between Europe and North America drop off the continental shelf into deep water [2].
This is the location of several major shipping routes [3] and will also cause an air exclusion zone between 0400-1400 each day.
Combined with suspected interference with similar cables off the north-west coast in August 2021 by a Russian Oceanographic Survey ship [3] there is potential for major interference or disruption to multiple cables simultaneously.
[0] https://liveuamap.com/
[1] https://www.navalnews.com/naval-news/2022/01/russian-navy-live-firing-off-irish-coast-during-tensions/
[2] https://world3dmap.com/submarine-cable-map/
[3] https://www.navalnews.com/naval-news/2021/08/russian-spy-ship-yantar-loitering-near-trans-atlantic-internet-cables/
8 comments
[ 3.8 ms ] story [ 75.2 ms ] threadSecondly is the affect on 'cloud' services (AWS, Azure, Goggle) with their complex inter-dependencies across services and geographical regions. We've seen several instances over the last couple of years where even a minor failure in one system has a knock-on effect to tens of thousands of other organisations that rely on it. Also, consider the reliance that AWS globally appears to have on US-East. If AWS Europe data-centers loose their link to US-East there could be major disruption of many services hosted on AWS, even those that use multiple availability zones.
Third would be the knock-on effect of switching to other routes (BGP) and the consequent 'traffic jam' as traffic between North America and Europe attempts to use alternate routes - maybe via Latin America and Africa or Asia.
Fourth, telecommunications ('phone) and VoIP services linking North America and Europe, Africa, and Middle East.
Fifth, and possibly the hardest to mitigate, is the by-product due to extensive use of CDN 'edge' services (Cloudflare, etc.). Most web sites nowadays have many dependencies on externally hosted services. Many are behind CDN edge servers. If the origin servers for those are on the other side of a major break then suddenly services you'd expect to be fully local will break in unexpected ways.
[0] https://www.swift.com/