Virtual Cross-Flow Detouring in the Deterministic Network Calculus Analysis

Deterministic Network Calculus (DNC) is commonly used to compute bounds on the worst-case communication delay in data networks. It provides various analyses to derive these bounds from a model, giving different tradeoffs between accuracy and analysis efficiency. Improving the tradeoff led to increasingly complex algorithms. We set out to design a novel DNC algorithm that is of low complexity while still providing competitive delay bounds. To achieve this goal, we make use of the insight that added pessimism in the model can alleviate more severe limitations of the DNC analysis. To that end, we introduce the concept of virtual cross-flow detouring where data flows are assumed to cross additional servers. Ultimately, we provide a heuristic that is simple, fast and high-quality. We show in numerical evaluations that our detouring not only provides a competitive alternative, it also outperforms current algebraic algorithms’ delay bounds for >50% of analyzed flows.

[1]  Fabien Geyer,et al.  On the Robustness of Deep Learning-predicted Contention Models for Network Calculus , 2020, 2020 IEEE Symposium on Computers and Communications (ISCC).

[2]  Jean-Yves Le Boudec,et al.  Network Calculus: A Theory of Deterministic Queuing Systems for the Internet , 2001 .

[3]  Jens B. Schmitt,et al.  Improving Cross-Traffic Bounds in Feed-Forward Networks - There is a Job for Everyone , 2016, MMB/DFT.

[4]  Eric Thierry,et al.  Tight performance bounds in the worst-case analysis of feed-forward networks , 2010, 2010 Proceedings IEEE INFOCOM.

[5]  Justine Sherry,et al.  Silo: Predictable Message Latency in the Cloud , 2015, Comput. Commun. Rev..

[6]  Ivan Martinovic,et al.  Improving Performance Bounds in Feed-Forward Networks by Paying Multiplexing Only Once , 2008, MMB.

[7]  Steffen Bondorf.  Better bounds by worse assumptions — Improving network calculus accuracy by adding pessimism to the network model , 2017, 2017 IEEE International Conference on Communications (ICC).

[8]  Ahlem Mifdaoui,et al.  Beyond the Accuracy-Complexity Tradeoffs of CompositionalAnalyses using Network Calculus for Complex Networks , 2017 .

[9]  Jörg Liebeherr,et al.  Fast Min-plus Convolution and Deconvolution on GPUs , 2017, VALUETOOLS.

[10]  Jens B. Schmitt,et al.  Catching Corner Cases in Network Calculus - Flow Segregation Can Improve Accuracy , 2018, MMB.

[11]  Jens B. Schmitt,et al.  Quality and Cost of Deterministic Network Calculus: Design and Evaluation of an Accurate and Fast Analysis , 2016, SIGMETRICS.

[12]  Luciano Lenzini,et al.  Estimating the worst-case delay in FIFO tandems using network calculus , 2008, VALUETOOLS.

[13]  Jens B. Schmitt,et al.  Boosting sensor network calculus by thoroughly bounding cross-traffic , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[14]  Fabien Geyer,et al.  DeepTMA: Predicting Effective Contention Models for Network Calculus using Graph Neural Networks , 2019, IEEE INFOCOM 2019 - IEEE Conference on Computer Communications.

[15]  Jens B. Schmitt,et al.  Calculating Accurate End-to-End Delay Bounds - You Better Know Your Cross-Traffic , 2016, EAI Endorsed Trans. Ubiquitous Environ..

[16]  Jens B. Schmitt,et al.  Delay Bounds under Arbitrary Multiplexing: When Network Calculus Leaves You in the Lurch... , 2008, IEEE INFOCOM 2008 - The 27th Conference on Computer Communications.

[17]  Alexander Scheffler,et al.  The Deterministic Network Calculus Analysis: Reliability Insights and Performance Improvements , 2018, 2018 IEEE 23rd International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD).

[18]  Marc Boyer,et al.  Deterministic Network Calculus , 2018 .

[19]  Georg Carle,et al.  Network engineering for real-time networks: comparison of automotive and aeronautic industries approaches , 2016, IEEE Communications Magazine.

[20]  Jens B. Schmitt,et al.  The DiscoDNC v2 - A Comprehensive Tool for Deterministic Network Calculus , 2014, VALUETOOLS.

[21]  Fabien Geyer,et al.  Performance Evaluation of Network Topologies using Graph-Based Deep Learning , 2017, VALUETOOLS.

[22]  Fabien Geyer,et al.  DeepComNet: Performance evaluation of network topologies using graph-based deep learning , 2019, Perform. Evaluation.