The exponential growth in data traffic is serving as a catalyst for the public network to evolve from a static, voice-centric infrastructure to a dynamic infrastructure that can respond to the shifts and changes of the new networked economy.
According to a recent study by RHK, the demand for
bandwidth from global networks could soar 300-fold in the next 8-10 years. Service providers are rushing to take advantage of new business opportunities resulting from this dramatic market shift. In this new and highly competitive marketplace, successful service providers will be those who differentiate themselves from the competition through unique service offerings.
At the forefront of this mandate for competitive differentiation is Intelligent Optical
Mesh, a new network architecture built on intelligent optical switching technologies that can fundamentally change the Service Provider business model and create a wide range of service delivery and application possibilities.
To better understand the business impact of a mesh network and what it will mean to the service provider market, we will explore the benefits of mesh and how these new networks compare to legacy SONET/SDH networks.
The Network Challenge
Contrary to popular belief, the challenge facing today’s networks is not how to create more capacity, it is how to transform capacity into usable bandwidth that can be exploited and packaged for customized service delivery.
Why are legacy networks incapable of delivering usable bandwidth?
The legacy or SONET/SDH network architecture was originally designed to transport voice traffic, not high-speed data network services. Voice traffic is predictable and characterized by slow growth. In contrast, data traffic is unpredictable and characterized by rapid growth. In order to quickly scale to meet this unpredictable demand for bandwidth, carriers must be able to cost effectively and rapidly deploy data networks.
Legacy networks cannot meet these newly realized data requirements and cannot manage the rapid growth in data traffic. Long, convoluted deployment cycles; costly and unnecessary protection allocation; and inflexible, long-term contracts and service level agreements (SLAs) locking customers into commitments they don’t need are just a few of the issues precluding legacy infrastructures from satisfying the complex needs of this new data-centric economy.
Adding Intelligence to the Public Network
In contrast to legacy architectures, a mesh-based network is designed for the unique needs of an unpredictable, data-centric market. Where legacy networks are hardware-centric and rigid, mesh networks are software-centric and offer an unprecedented level of flexibility.
In a mesh architecture, every element in the network is connected to every other element through an intelligent software-based network operating system (NOS). This means that communication between nodes can occur via many diverse routes. By adding this level of intelligence to the network, mesh enables rapid provisioning, rerouting and restoration of lightpaths automatically, without physical intervention or the need to reserve additional capacity for protection purposes. This greatly simplifies service provisioning.
Conversely, legacy networks are essentially point-to-point collections of “”dumb pipes”” that consist of a working route and a protection route. Services must be provisioned manually, requiring engineering resources to travel to the location of each network element and physically set up a circuit for service. This usually takes between 30 to 90 days depending upon the service.
In a mesh network, the process for service provisioning is simple and takes minutes.
A network operator selects a service route through several clicks of a mouse and a circuit is set up instantly through the intelligent routing and signaling software that exists on every element of the network. Time-to-revenue for the service provider is almost instantaneous. In an uncertain market, this sends a powerful message to customers as well as an increasingly skeptical financial community.
Customizing Bandwidth for Maximum Competitive Differentiation
Now that we have explored the differentiated, software-centric approach of a mesh network, we can delve into how this new architecture can enable customized service offerings.
Due to the flexibility of the mesh architecture, a network operator has the ability to allocate only as much working or protection bandwidth as customers require to meet current demand conditions.Service providers only pay for the service they need when they need it. Furthermore, these services are customized to meet their specific needs.
For example, a service provider can deliver 100 percent protection to its customers in the banking and financial service markets where critical applications require the highest levels of reliability. However, this same service provider can also offer a very inexpensive service with little or no protection to customers for non-critical applications such as Internet browsing. Additionally, this service provider can add various levels of restoration to further customize a service for customers with applications that fall in between a “”mission-critical”” and “”best-effort”” status.
In a legacy network, high-speed data service options are limited to one — Point to Point,
50 millisecond protected Private Line Service (OC-3, OC-12, OC-48). This service is slow to provision and upgrade and does not offer customization capabilities to address the unique needs of today’s data-centric services and high-bandwidth applications.
Additionally, a service provider is locked into bandwidth and protection schemes for the duration of the contract regardless of changing business demands and opportunities for revenue.
A mesh network does not lock a service provider into a specific business scenario.
Mesh provides the flexibility for service providers to take advantage of opportunities as they occur.
To further demonstrate this level of flexibility and customization, mesh services are comprised of two pricing components: port interface charges and virtual circuit charges.
A Web hosting company may purchase an OC-3 Virtual Private Network (VPN) service between five regional hosting centers connected via OC-12 ports at each center. Since the port interface speed (OC-12) is faster than the virtual circuit speed (OC-3), the Web hosting company would have the flexibility to increase the virtual circuit speed up to the port interface speed to meet variable traffic demands. The hosting company would also have the option to self-provision this additional capacity as needs warrant.