Calculadora Cables Ubicaciones Salud Investigación
← Todos los artículos
Análisis de ruta

Why Landlocked Countries Have Terrible Internet: The Physics of Connectivity

Why Landlocked Countries Have Terrible Internet: The Physics of Connectivity

Based on real RIPE Atlas measurements from GeoCables monitoring infrastructure, March 2026 Kazakhstan, Belarus, Bolivia, Ethiopia, Afghanistan, Switzerland. These countries share one characteristic that profoundly shapes their internet connectivity: no coastline. In a world where 95% of international internet traffic travels through submarine cables on the ocean floor, being landlocked means depending entirely on your neighbors for international internet access. The consequences show up clearly in our measurement data.

The Fundamental Problem

Submarine cables land on coasts. Landlocked countries cannot connect directly to submarine cables — they must route their international traffic through one or more coastal neighbors, adding terrestrial fiber segments, additional routing hops, and often monopoly pricing for transit. The physics are straightforward: - Light in fiber travels at ~200,000 km/s - Every 1,000km of terrestrial distance adds ~5ms of latency - Every network handoff (peering point) adds 1–5ms of processing delay - More transit operators = more cost = often less bandwidth available For a landlocked country in Central Asia, this means routing through Russia, China, or multiple Central Asian neighbors before reaching any submarine cable — adding 20–80ms before the international packet even reaches the ocean.

Kazakhstan: Our Primary Data Point

Our RIPE Atlas probe in Almaty, Kazakhstan provides the clearest picture of landlocked connectivity challenges. Kazakhstan is surrounded by Russia, China, Kyrgyzstan, Uzbekistan, Turkmenistan, and the landlocked Caspian Sea — no ocean access whatsoever. Kazakhstan → Indonesia (334ms): Our most-measured route from Kazakhstan shows the characteristic landlocked penalty. The traceroute from Almaty goes: Almaty → Astana (KZ) → London (RETN, AS9002, 44ms) → Frankfurt (NTT, 85ms) → Paris → Ashburn → San Jose → Osaka → Indonesia. The packet travels to London before heading east to Asia — going west first to reach a country that is east of Kazakhstan. This is the landlocked routing paradox: the nearest submarine cables are in China or the Persian Gulf, but BGP routing sends traffic to European hubs first because that's where the peering agreements are. Kazakhstan → Japan (344ms): Similar path — west to London, then NTT across the Atlantic and Pacific. Theoretical minimum via direct Central Asian route: ~120ms. Actual: 344ms. Overhead factor: 2.9×. Kazakhstan → Fiji (408ms): Our highest-latency Kazakhstan measurement. The packet crosses the entire globe: Almaty → London → Sydney → Suva. Kazakhstan to Fiji is 10,500km by great circle — the actual packet travels approximately 32,000km.

Belarus: The European Landlocked Case

Belarus presents a different variant of the landlocked problem. Geographically in Eastern Europe, surrounded by Russia, Ukraine, Poland, Lithuania, and Latvia — it has no coastline but is surrounded by countries with good submarine cable access. Belarus → South Korea (200ms): Our fastest long-distance route from Belarus goes north to Tallinn (Estonia, which has Baltic Sea access) then west to Frankfurt, then NTT across the Atlantic and Pacific. Tallinn — a coastal city — is the first point where Belarusian traffic reaches a network with submarine cable access. Belarus → China (181ms): Surprisingly efficient — Minsk → Frankfurt → Guangzhou via China Unicom's trans-Eurasian backbone. This works because the terrestrial fiber route across Russia is competitive with submarine alternatives for this specific city pair. Belarus → New Zealand (311ms): After Moscow (48ms), the next visible hop is Auckland (311ms) — 263ms of invisible submarine cable journey. Belarus depends entirely on Russian fiber to reach the Pacific.

The Transit Dependency Problem

Beyond latency, landlocked countries face a structural economic problem: transit dependency. A coastal country can connect directly to multiple submarine cable operators, creating competition that drives down bandwidth costs. A landlocked country must purchase transit from neighboring countries, which often means: Single-supplier risk: If the transit country has only one or two international cable providers, the landlocked country inherits that lack of competition. Kazakhstan's international traffic flows primarily through Russian networks (TransTeleCom, Rostelecom) — giving Russian operators significant leverage over Central Asian internet economics. Geopolitical exposure: When Russia restricted internet traffic in response to sanctions, Central Asian countries found their international connectivity impacted by decisions made in Moscow. Landlocked countries cannot easily build alternative routes without the cooperation of multiple transit nations. Pricing: Transit bandwidth typically costs 2–5× more than direct submarine cable access. This cost is passed to end users as higher retail internet prices.

The Solutions Being Tried

Trans-Caspian fiber: Kazakhstan and other Central Asian countries have invested in fiber cables across the Caspian Sea to Azerbaijan, which has both Caspian and Black Sea access, connecting to European networks via Georgia and Turkey. This reduces Russian transit dependency but adds Azerbaijan as a transit dependency. China as alternative transit: The Belt and Road Initiative's digital component has brought Chinese fiber investment to Central Asia. Our measurements confirm this works efficiently — Belarus → China via Frankfurt at 181ms, and the route shows China Unicom's direct European peering. But this substitutes Chinese transit dependency for Russian. Satellite: Starlink and other LEO satellite services are increasingly viable for landlocked countries as a supplement to terrestrial fiber. Latency is higher (~20-40ms) than fiber for regional routes, but provides independence from terrestrial transit.

Landlocked vs. Island: The Other Extreme

Ironically, island nations face a mirror problem. While landlocked countries can build terrestrial fiber cheaply but lack submarine cable access, island nations are completely dependent on submarine cables — often just one or two — with no terrestrial backup. Tonga, the Maldives, and many Pacific island nations have experienced complete international internet outages when their single submarine cable was damaged. Landlocked countries can at least reach multiple neighbors over terrestrial fiber; islands have no such option.

What the Data Tells Us

Across our measurement database, the landlocked penalty is consistent and measurable:
CountryBest measured RTTTheoretical minimumOverhead
Kazakhstan (Almaty)181ms (to China)~90ms2.0×
Belarus (Minsk)181ms (to China)~95ms1.9×
Kazakhstan (to Japan)344ms~120ms2.9×
Belarus (to Japan)273ms~100ms2.7×
Coastal countries at similar distances from the same destinations typically achieve 1.3–1.6× overhead — compared to 2.0–3.0× for landlocked nations. The gap represents real economic cost: slower cloud services, worse video call quality, higher latency for financial transactions, and reduced competitiveness for digital businesses compared to coastal counterparts.
See our full routing analysis for landlocked countries: Kazakhstan → Indonesia → · Belarus → Japan → · Belarus → Korea →