Thomas Levy

Embedded BGP Routing Monitoring

Previous studies ([7],[12]) explained a significant part of the BGP routing table growth in the Internet by the growth of specific operator practices like multi- homing and load-balancing. This article describes an embedded routing monitoring module aiming at detecting those practices, with the objective to help the router operator to better control BGP routing table size. The document first presents the motivations and the expected results of the study. Architecture of the routing monitoring module which has been prototyped are detailed, and results of experimentations are presented and analyzed. The article concludes with benefits and limitations of this approach, as well as ways this monitoring could be used to globally reduce BGP routing table thanks to route filtering.

A global study about Charging Correlation

The 3GPP standard body has defined the Online Charging System (OCS) as the central point in the charging architecture and Diameter Ro as the standard interface for enabling online charging in todays converged IP networks. It results in more and more heterogeneous telecom services charged by the Online Charging System. On top of that, operators want to deliver a consistent rating policy and aggregated billing information to their customers. As a result, there are emerging requirements for correlating lots of charging requests coming from the network into few elements rated at OCS level. This correlation of several charging requests (called “Charging Correlation”) is currently identified in various documents (such as 3GPP TS32.240 [1] or TS32.296 [2]). However, up to now, it has never been addressed in detail. For this reason, this document presents the various use cases where “charging correlation” is needed in IMS (IP Multimedia Subsystem), EPC (Evolved Packet Core) and Wimax context. Then, the document describes the impact of the use cases at network level and at Online Charging System level.