Feature Topic: Monitoring and Troubleshooting Multi-Domain Networks Using Measurement Federations

Post a reply

Natasha Simonovski

Feature Topic: Monitoring and Troubleshooting Multi-Domain Networks Using Measurement Federations

Thu, 10/31/2013 - 09:48 — Natasha Simonovski

In both the scientific and corporate worlds, users, resources, and data are often physically distributed, making networks increasingly important for all operations. Enormous progress has been made in increasing the capacity and accessibility of networking infrastructures, which in turn has fostered wider adoption of Cloud and Grid environments. Unfortunately, these advances have not directly translated into improved performance for all applications and users; instead, network performance problems become even more subtle and detrimental as the capacity of the network increases, and troubleshooting them on multi-domain network paths is highly challenging. These problems may be as benign as congestion from other network users, or as serious as packet loss caused by one or more intermediate-domain infrastructure and architectural flaws.

Troubleshooting performance problems on multi-domain networks requires a great deal of effort and expertise, as well as measurement policy agreements that mutually benefit domains within measurement federations. Novel approaches are needed to foster wider adoption of explicit measurement federations such as perfSONAR, SamKnows, Grenouille and M-Lab involving co-operating agents in collaborating vendor organizations as well as user communities. These approaches may also be suitable for implicit measurement federations seen in content-delivery networks involving multiple service providers that co-operate to reduce operating costs, while providing satisfactory end-user experience. Building upon current end-to-end measurement federation related standards-development efforts - at Open Grid Forum (OGF), IETF IP Performance Metrics (IPPM), IEEE 802.1 ag, ITU-T Y.1731, and Metro Ethernet Forum (MEF) - can benefit the interoperability and sustainability of explicit and implicit measurement federations.

In addition, sophisticated tools are required to monitor multi-domain networks and to detect, localize and diagnose performance problems in real-time. As networks increase in capacity, and new paradigms such as Software Defined Networking emerge to aid in traffic management, performance monitoring tools must be scalable and capable of detecting performance issues in a timely manner. The monitoring and diagnosing tools must comply with measurement federation policies, and aid network operators when troubleshooting perceived abnormalities, as well as help network middleware and intelligent applications to work around problems, ultimately minimizing the impact to end users.

 

Post a reply