NetView: Towards on-demand network-wide telemetry in the data center
Yunsenxiao Lin,Yu Zhou,Zhengzheng Liu,Ke Liu,Yangyang Wang,Mingwei Xu,Jun Bi,Ying Liu,Jianping Wu
DOI: https://doi.org/10.1016/j.comnet.2020.107386
IF: 5.493
2020-10-01
Computer Networks
Abstract:<p>Network telemetry is to collect information (<em>e.g.</em>, hop latency, throughput) from network devices. Network-wide telemetry is critical for operators to understand the quality of network performance and to diagnose on-going failures. The state-of-the-art telemetry approaches are far from ideal as they are unable to fully satisfy diverse requirements of operators, specifically for <em>on-demand, full coverage</em>, and <em>scalable</em> telemetry.</p><p>In this paper, we provide a new framework of network telemetry for data center networks, called <em>NetView</em>. NetView can support various telemetry applications and telemetry frequencies <em>on demand</em>, monitoring each device via proactively sending dedicated probes while only one vantage server is required. Technically, NetView divides the probe into a forwarding stack and a telemetry stack, which are respectively responsible for flexible forwarding and network status monitoring, achieving full <em>coverage</em> and visibility. Besides, a series of probe generation algorithms and update algorithms largely reduce probe number, providing high <em>scalability</em>. The evaluation shows that NetView reduces the bandwidth occupancy by more than two orders of magnitude compared with Pingmesh and INT-path, and conducts network-wide telemetry for the data center network with thousands of switches using only one vantage server, without bringing about resources bottleneck.</p>
computer science, information systems,telecommunications,engineering, electrical & electronic, hardware & architecture