CAIDA : tools : measurement : mantra

Mantra - Monitoring Multicast on a Global Scale

Mantra (Monitor and Analysis of Traffic in Multicast Routers) is a tool for monitoring various aspects of multicast at the router level. The results from Mantra are aimed to depict the snapshots of various constituents of the multicast infrastructure from the point of view of the router/s that we are monitoring. These snapshots, presented in several different modes, form the basis (and the main objective!) of this site. Results presented here are updated every 15-30 minutes.

Mantra was written by Prashant Rajvaidya as part of an ongoing project in Kevin Almeroth's Networking and Multimedia Systems Lab at the UC Santa Barbara Computer Science Department. Special thanks to Kevin Almeroth (UCSB), k claffy (CAIDA), David Meyer (Cisco) and Bill Fenner (AT&T Labs)

| General Monitoring Statistics Protocol Monitoring Interactive Visualizations Miscellaneous Site Tree |

cooperative project of:

NOTE: Mantra has not been actively maintained since 2003. The information regarding Mantra is mirrored here for historical purposes.

A list of publications related to Mantra is available in the section Mantra Publications

If you are interested in participating in Mantra please look at the section Participation in Mantra

		General Monitoring Statistics:
Status of Multicast Tables for a quick look at the latest statistics and the Top 10 Lists ; i.e. "most populous groups", "busiest hosts", "RP's advertising most SA's" etc. Session and Membership Monitoring Statistics about the usage of multicast and the membership patterns across entire multicast infrastructure visible to Mantra. mroute and Traffic Monitor Mroute monitoring involves analyzing characteristics of multicast routing table. On the other hand, traffic monitoring involves measuring the bandwidth usage and packet flow statistics for traffic through the routers that we monitor.	Multicast Usage in Last 24 Hours
		Protocol Monitoring:
MBGP Monitor Statistics about the characteristics of multicast routing. MBGP state captured from various routers form the basis of analysis here. MSDP Monitor Statistics about the characteristics of the MSDP-SA (Source Active) messages. Due to the *Same-SA-State-Everywhere* nature of MSDP , these results (theoretically) delineate Global usage statistics for multicast. Usage of GLOP Address Space Statistics about the usage of multicast-address space in *233/8* . GLOP is the multicast-address allocation policy for allocating a unique 233/24 address block to each AS. Usage of SSM Address Space Statistics about the usage of multicast-address space in *232/8* . Each AS is allocated a unique 232/24 address block that it can use for SSM (Single Source Multicast)	Multicast Routing in Last 24 Hours
Interactive Visualizations:
Topology-Views Visualization of local as well as global views of the MBGP topology. These visualizations also include display of monitoring statistics like the origin of MSDP messages are topological locations of session participants over the topology maps. Some interesting archived snapshots of the topology are also available in this section. Geographic Monitoring Interactive visualization of the geographic placement of important constituents of multicast infrastructure like participant-hosts, MBGP networks, DVMRP networks, RP's etc. The views presented here are not current and the IP-to-Geo mapping is not 100% accurate. However, these views present a very good estimate for multicast deployment across the globe.	A Snapshot of the MBGP Topology
Miscellaneous:
Routers that we Monitor Results specific to the routers that we collect data from. Mantra Publications List of publications related to Mantra. Other Tools A compendium of information about various other tools for monitoring different aspects of multicast. Participation in Mantra A semi-FAQ for those interested in participating in Mantra's monitoring efforts. Site Tree Site tree for Mantra-WWW.	Other Tools for Monitoring Multicast

Acknowledgements:

Mantra is an ongoing project in collaboration with the folks at CAIDA and Networking and Multimedia Systems Lab at the UC Santa Barbara Computer Science Department . Special thanks to Kevin Almeroth (UCSB), k claffy (caida) , David Meyer (Cisco) and Bill Fenner (AT&T Labs)

Comments and Suggestions:
Prashant Rajvaidya
<prash@caida.org>