4,700 km · 6 Landing Points · 3 Countries · Ready for Service: 2020
| Length | 4,700 km |
|---|---|
| Status | In Service |
| Ready for Service | 2020 |
| Landing Points | 6 |
| Countries | 3 |
| Location |
|---|
| Auki, Solomon Islands |
| Honiara, Solomon Islands |
| Noro, Solomon Islands |
| Port Moresby, Papua New Guinea |
| Sydney, NSW, Australia |
| Taro, Solomon Islands |
The Coral Sea Cable System (CS²) is a 4,700-kilometre submarine cable funded almost entirely by the Australian government that connects Sydney to Port Moresby in Papua New Guinea and to four landings in the Solomon Islands. It was commissioned in 2019, went live in 2020, and is now the primary international internet backbone for both small Pacific nations. The cable has a complicated origin story — it was built specifically to prevent Huawei from building the same cable — and our measurements reveal that, even now, its asymmetric use tells us something about which side of the link actually relies on it.
| Specification | Value |
|---|---|
| Length | 4,700 km |
| Ready for service | 2020 |
| Landings | Sydney, Port Moresby, Auki, Honiara, Noro, Taro |
| Owners | PNG DataCo Limited, Solomon Islands Submarine Cable Company |
| Funded by | Australian Department of Foreign Affairs and Trade (DFAT), ~AUD 137M |
In 2018, Huawei Marine Networks had signed preliminary agreements to build what would become the Coral Sea Cable System. The Solomon Islands government had negotiated with Huawei directly; Papua New Guinea was considering a parallel agreement. Australia — for whom PNG and Solomon Islands are the nearest Pacific neighbours and a long-standing strategic interest — intervened. Canberra offered to fund the cable from its own foreign-aid budget, on the condition that Huawei was not involved. The Solomon Islands government and PNG DataCo accepted, and Alcatel Submarine Networks (ASN, French-headquartered) became the supplier. The cable went into the Coral Sea in 2019 and opened for traffic in 2020.
The strategic logic was explicit. Australian security officials did not want a Chinese-state-aligned company to control the physical submarine cable linking Australia's closest Pacific neighbours to the global internet. The cable landings at Port Moresby and at Auki, Honiara, Noro, and Taro in the Solomon Islands were deliberately spread across the archipelago to maximise resilience: each major population centre reached by the cable could independently route around a fault at any single landing. Four Solomon Islands landings for a country with fewer than 700,000 people is lavish by normal commercial standards; it was not driven by commercial demand.
We monitor CS² in both directions between Sydney and Port Moresby, the cable's full southwest-northeast traversal. The data over 30 days is strikingly asymmetric:
| Direction | Samples | Min RTT | Avg RTT | Max RTT | Baseline | Ratio |
|---|---|---|---|---|---|---|
| Port Moresby → Sydney | 17 | 38.4 ms | 38.5 ms | 38.6 ms | 38.5 ms | 1.00 |
| Sydney → Port Moresby | 40 | 285.7 ms | 318.4 ms | 390.8 ms | 286.1 ms | 0.88 |
Port Moresby → Sydney runs at a rock-solid 38.4 ms minimum, 38.5 ms average, essentially zero variance — the packet enters the cable at Port Moresby, travels 2,700 km along CS² fibre to Sydney, and our measurement is done. The physics floor for that path is 46 ms; the observed value is slightly below the floor, which means our landing-point estimation overcounts the fibre length by a small margin. Regardless, it is a clean, stable measurement: CS² is doing exactly what it was built to do for traffic leaving PNG.
The reverse direction is a completely different story. Sydney → Port Moresby shows a 286-ms minimum and a 318-ms average — a 7.5-to-8× multiple of the theoretical floor. The stable baseline of 286 ms means this is not a transient anomaly; it is the normal routing pattern. 286 ms corresponds to roughly 28,000 km of fibre — the length of a trip across the Pacific to Los Angeles or Hawaii and back. Australian outbound traffic to Port Moresby is, in a surprisingly large proportion, routed via the United States and back across the Pacific on another cable, rather than going directly south to north on CS².
Why? Commercial peering. Australian Tier-1 carriers like Telstra and Optus have long-standing peering relationships with US-based networks that include PNG telecom destinations as part of their default upstream. Those peering relationships are older and better-priced than direct peering with PNG DataCo via CS². When the Australian carrier's routing policy says "send traffic to PNG via our US upstream," the packet enters a transatlantic trunk at Sydney, reaches a US West Coast IXP, transits the United States, and then uses a different cable (potentially one of the Trans-Pacific systems like Hawaiki or Pacific Data Gateway) to eventually reach Port Moresby. The journey is dominated by commercial routing history, not by the CS² cable that Australia paid for.
The Port Moresby → Sydney direction shows CS² being used heavily and cleanly. PNG DataCo, the Papua New Guinea telecom monopoly, uses the cable for its own outbound international traffic: every website request from a PNG user to any international destination traverses CS². This makes sense — PNG DataCo owns a share of the cable and pays for its operation; using it is the economically optimal choice for PNG-originated traffic.
The Sydney → Port Moresby direction shows that CS² is underused for inbound traffic. This is structural, not a fault. The cable fulfils its strategic purpose — it keeps PNG and the Solomon Islands connected to a non-Chinese-state-aligned international backbone, and it provides sovereign capacity for the two countries. Whether Australian-originated traffic uses the cable is a secondary question: the primary purpose was to ensure that PNG and Solomon Islands had a cable at all, and that it was not Huawei's. On both counts CS² succeeds.
The cable's four Solomon Islands landings are an unusual feature. Auki (on Malaita, population 7,000), Honiara (the capital, population 85,000), Noro (New Georgia, population 5,000), and Taro (Choiseul, population 1,000) represent four of the archipelago's main population centres. Most submarine cables serving a small Pacific nation would land only in the capital; CS² instead runs a branched cable body across the archipelago, touching four landings within the Solomon Islands alone. This is a strategic choice rather than a commercial one: four landings provide resilience against cable damage from typhoons, anchor strikes, or geological activity at any single site.
The arrangement is expensive to maintain but reflects the cable's funder. The Australian government treats CS² as foreign aid as well as infrastructure, and the four-landing architecture embodies the aid's distributional intent: every major Solomon Islands island community has direct cable access, not just the capital.
Live latency data on the CS² cable page. For comparison see Hawaiki (a larger Pacific cable reaching the US and New Zealand, 2018) and APG (an Asian consortium cable with a very different funding model). Our measurements refresh every two hours.
| Status | ✓ Normal |
|---|---|
| RTT | 38.56 ms / base 38.52 ms |
| Last checked | 2026-04-18 20:31 |
Monitored using RIPE Atlas probes. Open monitoring →
Find the actual cable routing distance between any two cities
Open Calculator →