Welcome to the SIG-PMV meeting at Zurich

Agenda: Presentations and Abstract

Connectivity Fault Management (CFM) is a standardised framework that addresses performance monitoring and troubleshooting challenges on Layer-2 end-to-end services, which consist of Layer-2 segments provided by different Layer-2 service providers. The main challenge is the detection of performance degradation and circuit interrupts together with cause localisation using functions directly on Layer-2. Additionally there is the need for alarming, triggering of consequent actions and support for L2 path analysis from end systems. This talk presents the concept and building blocks of CFM, shows its functionality in typical deployment scenarios and first results from a small physical test bed.

The use and need of crowd sourced based network performance measurement services and technologies is set to increase continually among the National Research and Education Networks (NRENs) in the near future. This requires an understanding of the behavior of network performance issues, and their localization and verification on wireless campus networks. The approach presented is based on the end-user mobile device measured feedback and allows the visualization of network performance in real time. Our thesis is - Is it possible to gather data from multiple sources, including browser-based measurements in addition to traditional monitoring, and extract meaningful information on the performance of a WiFi from that data?...Architecture building blocks, a walk-trough the system, and a real expertise with future steps shows our presentation.

Performance monitoring of optical/DWDM networks. This talk gives a presentation on PM on DWDM networks. The questions I will try to cover and discuss is: Which parameters to monitor, why should NRENs monitor the PM on DWDM links, and who could we do it and not least how it could help us in daily network operations. I will also show a demo on how we do DWDM-PM in our network. My intention is also to discuss how PM is done in different vendor platforms.

By sending probe packets with high frequency you can observe the dynamics of queue buildups and also measure dependability on a fine grained scale. By analyzing these outages we can get a clearer picture of the actual quality we deliver to our customers. Traditional measurements like OWAMP do not capture the fine grained problems. Even if we have redundant links and routes, the rerouting sometimes take much more time than necessary. Measurements are being done within UNINETT with 35 probes and also internationally with NORDUnet, Switch, RND(Brazil), and New Zealand.

RedIRIS new tool to identify performance problems in the institutions networks, is based on a server with HawkEye software. The user can perform tests to check connectivity quality that will be launched between clients (also called endpoints or probes) installed anywhere on their network. These clients are available for a variety of devices, from servers to mobile devices, bringing total flexibility. The actual set of tests include real services like Dropbox, Youtube, FTP or DNS. Also the institution can install multiple probes in different parts of the campus network, and launch tests between them, and even with probes from other institutions. There is total flexibility.

perfSONAR Small Nodes Project established an active network monitoring measurement platform of nearly 20 nodes covering GÉANT partners in Europe. The project was launched at TNC16 to enable new perfSONAR users in the GÉANT community to have firsthand experience on running a perfSONAR node and to introduce perfSONAR on low cost hardware. The measurement mesh platform runs latency and throughput measurements and is managed centrally by GÉANT project perfSONAR team. The authors will present the project overview, technical solution and the measurement platform, interesting performance troubleshooting case perfSONAR Small Nodes mesh helped resolve, and where the project is heading too

The preliminary AGENDA is in progress