The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Workshop:INDIS'21: 8th Workshop on Innovating the Network for Data-Intensive Science
Authors: Jeronimo Bezerra, Italo Brito, Arturo Quintana, Julio Ibarra, and Vasilka Chergarova (Florida International University); Renata Frez (Brazilian National Research and Educational Network (RNP)); Heidi Morgan (University of Southern California (USC)); and Marc LeClerc and Arun Paneri (NoviFlow Inc, Ontario)
Abstract: Long-haul networks are growing in complexity to address the constant need for more bandwidth, lower latency and jitter, customized traffic prioritization, and SLA-grade network resilience. A more complex infrastructure requires a deeper visibility of the assets to optimize the resource utilization as well as to protect the infrastructure and users connected to it. Leveraging legacy network monitoring technologies, such as SNMP, is not enough, since they do not offer real-time and granular visibility. That's where per-packet monitoring solutions can become a game changer. In-band Network Telemetry (INT) offers per-packet visibility with no impact to the network's forwarding plane. Adding per-packet visibility has the potential to change the network monitoring and operations field, and to redefine how traffic engineering will take place in the future. This paper aims to showcase how INT can dramatically increase network visibility, down to a sub-second scale, by comparing measurement results between INT and SNMP. Experiments and findings come from using the AmLight long-haul academic network as a use case.
Back to INDIS'21: 8th Workshop on Innovating the Network for Data-Intensive Science Archive Listing