Description

From 100GET-E3

Jump to: navigation, search

Key Data

Lifetime: 10/2007 - 09/2010
Total effort: 105,5 PYs
Total budget: 18.687,78 kEUR
Partners in Germany:
Partners in Finland:

Executive Summary

100GET-E3 devises an agile & TCO-efficient transport platform for IP & Ethernet services, based on throughput and capacity improvements, E2E control and management, as well as multi-layer optimization (MLO).

  • CAPEX-related investigations:
    • 100G packet processing (on-the-fly reconfigurability MPLS-TP <-> PBB-TE)
    • 100G long-haul transmission (coherent PDM-QPSK)
    • Multi-layer optimization of technological architecture (IP/MPLS & Ethernet/MPLS-TP over OTN over DWDM)
    • Multi-layer optimization in network planning
  • OPEX-related investigations:
    • Multi-layer & multi-domain optimization of functional architecture (management, control and path computation)
    • Multi-layer optimization in network operation (global reoptimization)
    • Demonstrations (GMPLS & PCE for DWDM, GMPLS interoperability via OIF UNI 2.0 & E-NNI)
    • Techno-economics of control plane automation and virtualization

Research topics

The ultimate goal of this research project is an agile & TCO-efficient transport platform that complements the existing IP (best effort) service with value-added (QoS-guaranteed) Ethernet services.

This not only improves individual customer experience but also shifts significant traffic volumes from the common IP/MPLS to a novel and more cost-efficient Ethernet/MPLS-TP switching platform.

On the CAPEX side, this goal translates into the following research topics:

  • 100G packet processing with on-the-fly reconfigurability between MPLS-TP & PBB-TE, increasing the throughput of packet nodes
  • 100G optical transmission & digital signal processing (DSP) for coherent PDM-QPSK optical long-haul interfaces, where the optical signal has a high spectral efficiency of 2 bit/sec/Hz and the symbol rate is reduced to 28Gbaud, making the required analog/digital conversion (ADC) components technically feasible and thus enabling the digital equalization of both CD and PMD effects. It can be used in existing fiber infrastructure without any group-velocity or polarisation-mode dispersion compensators - a key differentiating factor with respect to 50 Gbaud DQPSK and 100 Gbaud OOK transmission formats.
  • Streamlined layering of data plane technologies (IP/MPLS & Ethernet/MPLS-TP over OTN over DWDM)
  • Methods & algorithms for multi-layer optimized network planning, providing capacity resources in those amounts and locations as actually needed, taking into account constraints like available infrastructure, expected traffic and required fault tolerance.

Besides these CAPEX-related investigations, other aspects pertaining to OPEX are extensively studied, too. This integrated approach is important as in the near term, 100 Gb/s technology is deemed a solution for high-end applications in metro and long-haul transport networks. Being no mass product like Ethernet LAN technology, TCO-efficiency will only be achieved by means of multi-layer & multi-domain optimization of almost any operational aspect:

  • Streamlined functional architecture (management, control and path computation) across layers and domains, enabling end-to-end automation of resource discovery, tunnel provisioning & tunnel restoration. The planned resources are organized into a recursive hierarchy of tunnels with increasing granularity from MPLS(-TP) down to the DWDM layer, which efficiently grooms customers' end-to-end services like VPNs
  • Methods & algorithms for multi-layer optimized network operation, re-optimising the usage of the planned resources according to changing objectives like minimizing the maximum utilisation in busy hours, and maximizing the energy efficiency in idle periods. This effectively re-organizes the recursive hierarchy of tunnels according to the respective load conditions
  • Validation of operational concepts, like GMPLS & PCE applied to DWDM technology, or GMPLS compatibility in a multi-provider and multi-vendor environment via OIF UNI 2.0 & E-NNI interfaces
  • Techno-economical justification of the deployment of control plane or virtualization technologies, which significantly modify the business processes that are in use today
This page was last modified on 17 November 2009, at 09:41. This page has been accessed 7,648 times.