Hawkeye: Efficient In-band Network Telemetry with Hybrid Proactive-Passive Mechanism

With the rapid development of software-defined networking and programmable data plane, in-band network telemetry (INT) has attracted a lot of attention as an advanced network monitoring method. INT enables fine-grained and efficient network monitoring by encapsulating device status information into the packets. INT is divided into two types, proactive and passive, according to the source of the data packets carrying telemetry information. The proactive INT assembles large amounts of dedicated probe packets, with additional bandwidth overhead. The passive INT leverages the on-the-fly application flows, which saves the bandwidth cost but usually cannot achieve full coverage of network nodes. In this paper, we first propose a hybrid proactive-passive INT telemetry system, Hawkeye, which combines the proactive and passive INT methods together. Hawkeye leverages the advantages of the both methods and significantly saves the bandwidth cost while achieving telemetry coverage. To make it, Hawkeye divides the network topology into two parts, proactive INT network part and passive INT one, based on the historical traffic information. First, Hawkeye offloads passive telemetry tasks with the existing steady flows. Then, Hawkeye leverages a path planning algorithm, P2A, to generate the best possible probe paths to cover the rest of the network, which are not covered by passive telemetry. Simulation results show that Hawkeye can reduce the average bandwidth overheads by 55.2% and achieve 100% network link coverage for the real-world network topology, compared against the existing INT methods.

[1]  Wei Gao,et al.  Efficient Clustered Network Telemetry based on Failure Awareness , 2022, 2022 23rd Asia-Pacific Network Operations and Management Symposium (APNOMS).

[2]  Huiling Shi,et al.  A Packet Loss Monitoring System for In-Band Network Telemetry: Detection, Localization, Diagnosis and Recovery , 2021, IEEE Transactions on Network and Service Management.

[3]  Wei Su,et al.  In-band Network Telemetry Task Orchestration based on Multi-objective Optimization , 2021, 2021 22nd Asia-Pacific Network Operations and Management Symposium (APNOMS).

[4]  Zuqing Zhu,et al.  Highly-Efficient and Adaptive Network Monitoring: When INT Meets Segment Routing , 2021, IEEE Transactions on Network and Service Management.

[5]  Tian Pan,et al.  INT-probe: Lightweight In-band Network-Wide Telemetry with Stationary Probes , 2021, 2021 IEEE 41st International Conference on Distributed Computing Systems (ICDCS).

[6]  Na Li,et al.  In-band Network Telemetry: A Survey , 2021, Comput. Networks.

[7]  Zuqing Zhu,et al.  Sel-INT: A Runtime-Programmable Selective In-Band Network Telemetry System , 2020, IEEE Transactions on Network and Service Management.

[8]  Yangyang Wang,et al.  NetView: Towards On-Demand Network-Wide Telemetry in the Data Center , 2020, ICC 2020 - 2020 IEEE International Conference on Communications (ICC).

[9]  Arthur Francisco Lorenzon,et al.  Orchestrating In-Band Data Plane Telemetry With Machine Learning , 2019, IEEE Communications Letters.

[10]  L. Gaspary,et al.  An optimization-based approach for efficient network monitoring using in-band network telemetry , 2019, Journal of Internet Services and Applications.

[11]  Zuqing Zhu,et al.  Visualize Your IP-Over-Optical Network in Realtime: A P4-Based Flexible Multilayer In-Band Network Telemetry (ML-INT) System , 2019, IEEE Access.

[12]  Bin Liu,et al.  INT-path: Towards Optimal Path Planning for In-band Network-Wide Telemetry , 2019, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[13]  Sangheon Pack,et al.  Selective In-band Network Telemetry for Overhead Reduction , 2018, 2018 IEEE 7th International Conference on Cloud Networking (CloudNet).

[14]  Roberto Bifulco,et al.  A Survey on the Programmable Data Plane: Abstractions, Architectures, and Open Problems , 2018, 2018 IEEE 19th International Conference on High Performance Switching and Routing (HPSR).

[15]  Jennifer Rexford,et al.  HULA: Scalable Load Balancing Using Programmable Data Planes , 2016, SOSR.

[16]  Hua Chen,et al.  Pingmesh: A Large-Scale System for Data Center Network Latency Measurement and Analysis , 2015, SIGCOMM.

[17]  Thierry Turletti,et al.  A Survey of Software-Defined Networking: Past, Present, and Future of Programmable Networks , 2014, IEEE Communications Surveys & Tutorials.

[18]  George Varghese,et al.  P4: programming protocol-independent packet processors , 2013, CCRV.

[19]  George Varghese,et al.  Forwarding metamorphosis: fast programmable match-action processing in hardware for SDN , 2013, SIGCOMM.

[20]  Min Zhu,et al.  B4: experience with a globally-deployed software defined wan , 2013, SIGCOMM.

[21]  Jürgen Quittek,et al.  Requirements for IP Flow Information Export (IPFIX) , 2004, RFC.

[22]  Anja Feldmann,et al.  NetFlow: information loss or win? , 2002, IMW '02.

[23]  Li Chen,et al.  LightGuardian: A Full-Visibility, Lightweight, In-band Telemetry System Using Sketchlets , 2021, NSDI.

[24]  Anirudh Sivaraman,et al.  In-band Network Telemetry via Programmable Dataplanes , 2015 .

[25]  Jeffrey D. Case,et al.  Simple Network Management Protocol (SNMP) , 1990, RFC.

[26]  Carl A. Sunshine,et al.  Source routing in computer networks , 1977, CCRV.