Retirement at Age 25?

The centrality of our submarine cable system to global economics and communications cannot be understated. There is still a common public misunderstanding that satellites are the primary method of international communications, however submarine fiber optic cables now carry almost 100% of the world’s international Internet, voice and data communication.

Submarine cables are high-cost infrastructure projects that are difficult and expensive to repair after they have been deployed. When repair is required, the cost is high, and the out-of-service time can be weeks or even months.

The industry norm has been to engineer the submerged plant for a lifetime of 25 years. This means there will probably be (at most) one failure of the equipment during the first 25 years of life that requires a ship repair.

Undersea Evolution

Interestingly, the first submarine communications cables were laid in the 1850s. Since then, traffic that ran across the Atlantic Ocean as an electrical Morse code signal has been replaced with WDM optical telecommunications transmission technologies. Data rates of 10 Tb/s per cable are now achievable; a factor of about 1013 times higher than the telegraph transmissions over copper wire 150 years ago.

Subsea telecommunications cables are still laid, as they were for Morse code, in water depths of up to 8,000 m where the water pressure is approximately 8,000 Tonnes force per square metre. The design of the pressure vessel to contain the electronics, including the watertight seals where the cable and glass optical fibers enter and leave the repeater housing requires sophisticated engineering and specialised technologies.

Meeting these unique environmental demands is tough and there are only a handful of suppliers that have the capability to design, manufacture and deliver the undersea portion of these systems. Manufacturing component parts requires specialised clean-room environments, and uses ultra-high reliability components as well as duplicated or redundant electronics to achieve extremely high reliability.

The threats to these cables are not just environmental. Submarine cables are subject to external aggression from natural and human causes, and may also fail due to component or equipment malfunctions. The undersea route for the cable is chosen with care to avoid damage by external aggression as much as possible.

Despite best efforts, cable cuts and cable failures will happen; making the protection of the traffic circuits the most important issue for cable owners and telecom carriers.

Fortunately, new technologies have enabled us to provide coherent detection WDM 100 Gb/s traffic over legacy trans-Pacific submarine networks, allowing intelligent mesh protection for these spans. We can achieve traffic resiliency against failures due to external aggression or equipment malfunction, and treat terrestrial and subsea portions of networks as a single, global, network.

Is a 25-Year Design Life Still Appropriate?

A design lifetime of 25 years for the submerged portion of subsea cables has been the norm for many years. However in the 2000-2004 timeframe, submarine cable owners looked at the possibility of cables becoming full at around 10-15 years, and that newer cable with higher ultimate capacity might be more cost-effective. Submarine plant suppliers investigated the options, but concluded that a design lifetime of 12 years or 15 years still required the same ultra-high reliability components and the same duplicated or redundant circuitry. Cutting the life of a cable in half would save only a few precent of the overall cost.

The explosion in global bandwidth demand, combined with a new suite of coherent technologies has placed a renewed focus on the health and capacity of subsea cable.

The technologies for the submerged plant of submarine cable systems have not changed much over the past 10 years. However, in the past 5 years the technologies for optical communications has moved ahead quickly. The introduction of dense WDM technologies, phase modulation techniques, and in particular coherent detection 40 Gb/s and 100 Gb/s technologies to submarine line terminal equipment by suppliers such as Ciena has dramatically increased the capacity that can be achieved on existing submarine cables. In some cases an ultimate capacity of up to 80 times the original design capacity of the submerged plant has been achieved, simply by applying new SLTE to the ends of the cable. In recent years an increase of 20 to 40 times the original ultimate capacity has been achieved.

This trend of upgrading SLTE with the latest technology, and the resulting increase in ultimate capacity, has meant that a large part of the increased demand can be met by existing subsea cables, by upgrading only the land based SLTE terminal equipment.

If the technology is in place to meet bandwidth demands, the question becomes whether the existing cables can last. Is a 25-years design life long enough, or do cable owners now hope to extend the "retirement" of their submerged cable beyond this point? Will the cables actually last that long?

Some large submarine cable companies have already publicly announced that they now expect their submerged equipment to be practically and economically viable for at least 25 years (and perhaps 30 years) thanks to SLTE technology upgrades. Many of the submarine cables built between 10 and 15 years ago have indeed proven to be extremely reliable, and have had few problems in terms of external aggression or optical or electronic reliability.

Protection Against Multiple Modes of Failure

Whether service outages originate from ship anchors cutting the cable or from hardware malfunctioning, protection of the traffic circuits is critical. The same technology advances that have brought us 100 Gb/s traffic waves on trans-Pacific submarine networks also allow us to implement intelligent mesh protection in both subsea and terrestrial networks. We can now treat the network as a single global asset rather than two separate terrestrial and subsea portions.

Intelligent mesh networks are able to quickly and autonomously reroute connections around inevitable submarine cable faults, whether they be due to rogue anchors, an earthquake or the failure of a pump laser or integrated circuit within a submarine repeater.

Properly designed intelligent mesh topologies can achieve network availability that which is a tenfold improvement over the legacy ring-based networks used to protect submarine networks in the past. These architectures can enable operators to survive multiple simultaneous network faults, whether caused by external aggression or internal hardware faults.

Innovation in network architecture, combined with new transmission technologies adds up to a delayed retirement for the world’s deployed subsea cables. All of this could extend the life of deployed cables beyond the expected 25-year lifespan.

Colin Anderson is Ciena’s manager for business development and marketing for submarine networks. He has 25 years of experience in sales and marketing, engineering, and business development roles in the subsea and terrestrial telecommunications markets. Colin was Program Chairman for SubOptic 2010 held in Yokohama Japan in May 2010, and is a past member of the SubOptic Executive Committee. He is a Senior Member of the IEEE. Colin can be reached at[email protected]. For more information, please visit www.ciena.com.