
The increasing need for high-precision monitoring and network-wide observability has spurred significant interest in In-Band Network Telemetry (INT), which empowers the collection of monitoring data directly from the dataplane, thus, alleviating the various shortcomings of passive monitoring techniques. INT, in some of its initial forms (e.g., P4-INT), tends to accumulate per-hop telemetry values within each packet, introducing substantial transmission overhead. Recent efforts to alleviate this shortcoming spread per-hop telemetry values among multiple packets, but eventually cause other implications, such as INT data redundancy or collisions during state lookup, stemming from the probabilistic data structures that they employ. Along these lines, we stress on the need for a lightweight INT approach with deterministic behavior. To this end, we present a new INT approach (Controller-Assisted Lightweight In-Band Network Telemetry – CLINT) that relies on a telemetry controller for the coordination of INT nodes, through the update of telemetry states, which indicate at each instance the telemetry action that should be applied to each incoming packet. We discuss in detail the operation of CLINT, as well as its prototype implementation using P4 BMv2 and the P4Runtime API. Our evaluation results indicate that CLINT outperforms two state-of-the-art lightweight INT techniques (i.e., PINT and DLINT) in terms of monitoring efficiency and network update detection, while maintaining an insignificant control communication overhead.
network monitoring, In-band network telemetry (INT), Electrical engineering. Electronics. Nuclear engineering, software-defined networks, programmable dataplanes, TK1-9971
network monitoring, In-band network telemetry (INT), Electrical engineering. Electronics. Nuclear engineering, software-defined networks, programmable dataplanes, TK1-9971
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 1 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
