Next generation networks in need of redefinition

Stuart Corner
Tuesday, 03 June 2008 06:22

Business IT - Technology

A senior Juniper Networks engineer argues that the current distinction between transport and services in today's telecommunications networks is flawed and will lead to significant inefficiencies in future next generation networks unless a different approach is adopted.

According to Juniper Networks Fellow, Kireeti Kompella, the core of the problem is the role played by MPLS in today's networks. He argues that MPLS was developed to be a transport technology but has been co-opted to deliver services and now needs to be "pulled apart so that some functions remain in the transport layer and some in services." However the roots of the problem go back further and, he argues are so deeply entrenched that they impact every player in the industry.

In a presentation to Juniper's service provider customers at the company's J-Tech Asia Pacific Forum in Bangkok last week, Kompella said: "The 'elephant' in the room is the...line that separates the 'IP/MPLS' part of the network from 'transport'. This has become entrenched in network architectures over the past 20 years, and manifests itself: in organisational structures, both in service providers and equipment vendors; in philosophies of deployment and implementation; in regulations and laws set by governments; in unions and internal politics of various flavours; and in so many other hidden or unnoticed ways."

He argues that in the time division multiplexed networks of two decades ago - in which several independent services rode over the transport network - the line separating 'infrastructure' from 'services' made perfect sense, and keeping infrastructure separate allowed for a very stable network, over which each service could be managed on its own.

Today, he says "there is essentially just one 'service' over transport, namely IP/MPLS. The real services are further above, and since IP/MPLS carries all these services, it must have the same stability and resilience as the infrastructure below, and therefore must be part of the transport infrastructure."

Just how and where the line is drawn remains a matter for debate, but according to Kompella, the issue must be confronted. "We (as a total community) have to confront [the issue]. If not, one cannot build efficient, cost-effective networks....[and] 'next generation networks' [will be] just incremental changes, not real progress towards a packet-centric paradigm. All the talk of saving capex and opex will also be incremental, rather than fundamental changes towards an ultra-high-bandwidth future."
CONTINUED