Recent research results and ongoing work

Thu, 03/27/2008 - 20:02 by Olivier Bonaventure • Categories:

The IP Networking Lab of Université catholique de Louvain in Louvain-la-Neuve is actively working on solving problems that affect network engineers working for Internet Service Providers, large enterprise or router manufacturers. To allow companies to benefit from these research results, we organise from time to time half-day tutorial seminars that describe our recent results.

Our last seminar was organised on Thursday April 24th, 2008 in Louvain-la-Neuve. If you would like to be informed about upcoming seminars, please send an email to Olivier Bonaventure.
The slides and recordings for most of the presentations are available below.

Recent results

Modeling ISP networks with C-BGP

pdf slides, recording ipod, recording mp4

Large ISP networks heavily depend on the BGP protocol to provide Internet services, but also for BGP/MPLS VPNs. BGP is a powerful, but complex protocol and some configuration tweaks may significantly affect the performance of networks that are using BGP. In this talk, Bruno Quoitin will provide a detailed presentation of how the C-BGP network modelling tool can be used by network engineers to perform what-if scenarios in order to improve their BGP configuration.

Fast convergence and fast reroute in pure IP networks

pdf slides

Given the requirements form mission critical applications, more and more IP networks need to be able to recover quickly from link failures. In large networks, MPLS fast reroute techniques are often deployed to solve this problem. However, MPLS is not the only solution to this problem.
In this talk, Pierre François will discuss the techniques that can be used in pure IP networks to deal with planned and sudden topology changes. For the planned topology changes, the key issue is how to avoid losses of connectivity and transient loops during the OSPF/ISIS or BGP convergence that follows the change. The presentation will describe how metric increments can allow OSPF and ISIS to support graceful topology changed and how BGP policies can be modified when planned failures happen. For the sudden topology changes, the presentation will discuss both the recent improvements to the implementations of IP routing protocols as well as pure IP fast reroute techniques that are being developped within the IETF.

IPv6 multihoming with shim6

pdf slides, recording ipod, recording mp4

According to the latest predictions, IPv4 addresses will be exhausted by 2012. This gives a strong incentive to network operators and content providers to deploy IPv6 and provide commercial IPv6 service. Compared to IPv4, IPv6 gives a much large addressing space. This larger addressing space provides several benefits. One of them is that it allowed the IETF to design a more powerful and more scalable multihoming technique compared to the existing IPv4 multihoming hacks. In this presentation, Olivier Bonaventure will describe the current state of multihoming in IPv6. He will explain in more details the shim6 host-based IPv6 multihoming solution that is being finalised within the IETF and discuss the implications of this technique for enterprise networks and ISPs. In addition, Sébastien Barré will give a demonstration of the open-source shim6 implementation that he developped in the Linux kernel.

The LISP approach to improve Internet scalability

pdf slides, recording ipod, recording mp4

Very recent activities in the IETF and in the Routing Research Group (RRG) of the IRTF focus on defining a new Internet architecture, in order to solve scalability issues related to interdomain routing. The approach that is being explored is based on the separation of the end-systems’ addressing space (the identifiers) and the routing locators’ space. This separation is meant to alleviate the routing burden of the Default Free Zone, but it implies the need of distributing and storing mappings between identifiers and locators on caches placed on routers. In this talk, Luigi Iannone gives an overview of this new paradigm and explains both the benefits and the costs that it introduces in the Internet Architecture.

The case for an informed path selection service

pdf slides, recording ipod, recording mp4

During the past ten years, we have seen the emergence of a set of applications requiring more and more quality of service (QoS). For instance, videos need large bandwidth and delays as lows as possible. Further, while previously a content was located in a single place, it is, nowadays, frequent that the content is replicated among a set of servers located anywhere on five continents or even among users themselves. Perfect examples of this are peer-to-peer (P2P) applications. In addition, multihoming, i.e., the ability of having different connections to Internet potentially through different providers, is becoming more and more popular. Finally, network level protocols such as SHIM6 or LISP must often choose the best path among a list of highly disparate paths according to traffic engineering or policies considerations. This situation leads to more complexity in ensuring QoS to the user. We have now to maximize the QoS perceived by the user based on a set of highly disparate paths. In addition, in many new applications, path selection and QoS are ensured at the application layer resulting in an ineffective network level resources
use.

Today, although many of measurements techniques have been developed within the IPPM working group of the IETF, an application that needs to select a path or a server must implement its own measurement system to obtain data to perform its selection. Thus, several applications running on the same host or in the same campus will probably perform almost the same kind of measurements. Duplicating those measurements is not the appropriate solution. A better solution to this problems would be to develop a service that runs continuously and could be queried by applications requiring QoS and path information. In this talk, Benoit Donnet describes our proposed service, named IDIPS (for ISP-Driven Informed Path Selection ). We further present an evaluation of our current implementation.

Towards validated router configurations

pdf slides, recording ipod, recording mp4

Many networks are still configured manually on a router by router basis. This is errorprone and often leads to misconfiguration problems. In this talk, a different approach is proposed. Instead of configuring the routers individually, we developped a tool that allows the network operator to design efficiently correct network configurations. Our tool offers an easy way to check and generate router configurations based on a high-level view of a network. For each module, the network operator may specify preconditions that are automatically validated by the tool. So it acts as an alternative to the classical trial-and-error approach. Modularity, extensibility and the ease of use were the tool main priorities. We conclude the presentation with a relevant case study based on a real network (Abilene, the US research network) and explain how the corresponding Juniper configurations are generated.