The evolution in packet-optical transport systems

For many operators, early packet-optical transport system (P-OTS) installations failed to deliver on the promises of operational efficiency, capex savings, optimized total cost of ownership (TCO), and lower cost per bit due to a number of issues. Let’s explore the P-OTS offerings that have been developed to date – as well as their limitations – and how P-OTS technology should evolve to create a more viable “beyond NG” transport platform capable of easing operators’ scalability and profitability pain points.

Nov 23rd, 2011

The tremendous growth of data services over the last several years has resulted in unprecedented bandwidth and capacity challenges for operators. In response to these challenges, infrastructure vendors quickly introduced the packet-optical transport system (P-OTS), a next-generation (NG) optical transport platform designed to address the issues of bandwidth, capacity, and scale in telecom networks.

But for many operators, early P-OTS installations failed to deliver on the promises of operational efficiency, capex savings, optimized total cost of ownership (TCO), and lower cost per bit due to a number of issues. Let’s explore the P-OTS offerings that have been developed to date – as well as their limitations – and how P-OTS technology should evolve to create a more viable “beyond NG” transport platform capable of easing operators’ scalability and profitability pain points.

A look at P-OTS to date

Ever since the introduction of the P-OTS concept, infrastructure vendors have worked hard to develop platforms that aim to make today’s data-deluged transport networks more supportive of new services, more resilient, easier to manage, and more cost-effective. However, to date these systems have failed to live up to expectations due to a number of shortcomings or pitfalls. These include:

Minimal Layer 2 functionality: A number of vendors have taken the path of simply adding Layer 2 switching to a blade on their existing WDM systems. The problem with this approach is that it’s primarily focused on Layer 1 and these systems lack any scalable data switching or full Carrier Ethernet capabilities. While these vendors may position their offerings as end-to-end systems, in reality a different device is required for each layer in the network and, as a result, the full synergies of a true P OTS can never be achieved.

Single-layer management: Some vendors added P-OTS functionality just at the management layer. This approach is not designed to span the metro and core end-to-end within a single platform and thus can strand potential capex and opex savings.

Multiple devices supporting each layer: Other vendors require operators to purchase and deploy a variety of different shelves to get ODU-crossconnect or ROADM capabilities. This requirement reduces the operational and packaging efficiencies that a well-designed P-OTS should offer.

Overly integrated designs: Still other vendors have brought to market an inflexible mix of integrated technology that requires operators to invest in functionality they do not yet need. This results in a high-cost entry point and a great potential to strand capex. As well, many router vendors have chosen to implement full Layer 3 intelligence end to end, regardless of cost or need. This design strategy disregards the attractive and well-proven benefits of using traditional transport OAM functionality to embrace the packet layer.

What all these approaches have in common is that they involved repurposing existing platforms and concepts to meet time-to-market or budget objectives. A better approach would see a bona fide NG transport system built from the ground up to be capable of more cost effectively scaling and more efficiently supporting packet- and circuit-based services (and even converging networks when the time is right). Because of this, new requirements have emerged to address these challenges and transform the P-OTS into a more practical NG transport tool.

The evolution of P-OTS ‘beyond next generation’

While the P-OTS platforms of the past did not fully meet operators’ requirements for NG transport, they did succeed in demonstrating the value of integrating optical and packet technology in a single platform. The next step in the evolution of P-OTS clearly would be a purpose-built system that addressed the shortcomings we’ve just described.

Such a multi-layered transport system would marry the scalability, performance, cost, and operational abilities of optical networks with the adaptability, cost-effectiveness, and flexible service creation aspects of packet networks. Specifically, such a platform would provide:

  • Convergence of network layers to reduce the cost of building and operating networks
  • Integration of Layer 0-3 functionality in a single platform to reduce capex, opex, and cost per bit
  • A modular and flexible design that enables a low entry cost and pay as you grow scalability for added functionality
  • A single management system for all network layers, allowing new services to be introduced quickly and maintained easily.

Such a platform would offer true NG transport, as it would fully integrate multiple network layers in a modular, flexible design that enabled operators to scale cost-effectively, make maximum use of network resources, optimize TCO, and protect existing revenue streams. It also would enable the creation of new revenue opportunities.

The way forward
Although early generation P-OTS may have not achieved their full potential due to less than ideal execution, they thankfully offered a clear direction for evolution toward a more optimal approach. The next step in P-OTS will offer more intelligent, efficient, and flexible “beyond NG” transport. The resulting system should fully integrate the packet and optical layers to enable operators to minimize cost and maximize the profitability of services, both today and well into the future.

Jimmy Mizrahi is NG optical networking product line manager at ECI Telecom.

More in Network Design