Runtime reconfiguration of time-sensitive networking (TSN) schedules for Fog Computing

Fog Computing is about to tremendously impact the industrial automation industry. While, today's industrial automation systems use proprietary technology to provide real-time and dependability guarantees, Fog Computing enables a paradigm shift from this proprietary operations technology (OT) to the usage of standard IT equipment and infrastructure. Thus, Fog Computing is one of the key elements of the Industrial Internet of Things and Industry 4.0. In particular, future Fog Computing will be enabled by: the increased usage of IP-protocols, e.g., standardized Deterministic Ethernet solutions from IEEE Time-Sensitive Networking (TSN) Task Group, upcoming 5G wireless standards, and interoperability standards such as OPC Unified Architecture (OPC UA). In this paper, we address the runtime reconfiguration of TSN-based Fog Computing platforms. We assume that the Fog Computing platform is composed of several heterogeneous Fog Nodes interconnected using TSN. The applications consist of hard real-time control tasks, which have strict timing requirements expressed as hard deadlines. For time-critical traffic, TSN uses the scheduled traffic type, which relies on Gate-Control Lists (GCLs) at each outgoing port of a network switch to decide the transmission of scheduled frames. We propose a heuristic algorithm to determine the GCLs at runtime such that the deadlines are satisfied and the queue usage is minimized, to accommodate non-critical traffic. The scheduling algorithm is part of a configuration agent that reacts to network changes and reconfigures the GCLs using the NETCONF network configuration protocol. Our experimental evaluation shows that the scheduling heuristic is able to quickly find good-quality GCLs.

[1]  Silviu S. Craciunas,et al.  Design optimisation of cyber-physical distributed systems using IEEE time-sensitive networks , 2016, IET Cyper-Phys. Syst.: Theory & Appl..

[2]  Jiang Zhu,et al.  Fog Computing: A Platform for Internet of Things and Analytics , 2014, Big Data and Internet of Things.

[3]  Wilfried Steiner,et al.  An Evaluation of SMT-Based Schedule Synthesis for Time-Triggered Multi-hop Networks , 2010, 2010 31st IEEE Real-Time Systems Symposium.

[4]  Frank Dürr,et al.  No-wait Packet Scheduling for IEEE Time-sensitive Networks (TSN) , 2016, RTNS.

[5]  Sasikumar Punnekkat,et al.  A configuration agent based on the time-triggered paradigm for real-time networks , 2015, 2015 IEEE World Conference on Factory Communication Systems (WFCS).

[6]  Silviu S. Craciunas,et al.  Scheduling Real-Time Communication in IEEE 802.1Qbv Time Sensitive Networks , 2016, RTNS.

[7]  Stefan Poledna,et al.  Fog computing as enabler for the Industrial Internet of Things , 2016, Elektrotech. Informationstechnik.

[8]  Giorgio Buttazzo,et al.  Hard Real-Time Computing Systems: Predictable Scheduling Algorithms and Applications , 1997 .

[9]  Paul Pop,et al.  Design optimization of TTEthernet-based distributed real-time systems , 2014, Real-Time Systems.

[10]  Raja Lavanya,et al.  Fog Computing and Its Role in the Internet of Things , 2019, Advances in Computer and Electrical Engineering.

[11]  Silviu S. Craciunas,et al.  Combined task- and network-level scheduling for distributed time-triggered systems , 2016, Real-Time Systems.

[12]  Oliver Sinnen,et al.  Task Scheduling for Parallel Systems , 2007, Wiley series on parallel and distributed computing.

[13]  Theodore J. Williams,et al.  The Purdue Enterprise Reference Architecture , 1992, DIISM.