Adaptive industrial IOT/CPS messaging strategies for improved edge compute utility

Messaging strategies for data transport play a central role in large scale IIoT/CPS systems. Research to date in these domains has focused on comparative studies of different messaging protocols and various congestion control techniques. This paper presents an experimental evaluation of the latency and throughput for different message payload strategies using the same protocol combined with a novel congestion control algorithm. The results provide key insights based on experiments carried out in a real-world IOT wireless sensor network deployment. The approach presented dynamically varies the message schema and sizes in response to the utilization of edge nodes. The goal is to achieve the highest possible data throughput that fits within the available edge based compute capacity without requiring auto-scaling of back end cloud resources.

[1]  Alexander Gluhak,et al.  A survey on facilities for experimental internet of things research , 2011, IEEE Communications Magazine.

[2]  Lifeng Zhou,et al.  Industry 4.0: Towards future industrial opportunities and challenges , 2015, 2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD).

[3]  Claus Pahl,et al.  Workload Patterns for Quality-Driven Dynamic Cloud Service Configuration and Auto-Scaling , 2014, 2014 IEEE/ACM 7th International Conference on Utility and Cloud Computing.

[4]  Thomas Kunz,et al.  Performance evaluation of IoT protocols under a constrained wireless access network , 2016, 2016 International Conference on Selected Topics in Mobile & Wireless Networking (MoWNeT).

[5]  Omran Saleh,et al.  Monitoring and Autoscaling IaaS Clouds: A Case for Complex Event Processing on Data Streams , 2013, 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing.

[6]  Anna Larmo,et al.  Impact of CoAP and MQTT on NB-IoT System Performance † , 2018, Sensors.

[7]  Jesus Alonso-Zarate,et al.  A Survey on Application Layer Protocols for the Internet of Things , 2015 .

[8]  Michael Weyrich,et al.  Machine-to-Machine Communication , 2014, IEEE Software.

[9]  Marco Aurélio Stelmar Netto,et al.  Using Application Data for SLA-Aware Auto-scaling in Cloud Environments , 2015, 2015 IEEE 23rd International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems.

[10]  Tao Chen,et al.  Self-aware and self-adaptive autoscaling for cloud based services , 2016, ArXiv.

[11]  Yves Chauvin,et al.  Backpropagation: theory, architectures, and applications , 1995 .

[12]  Kaijie Zhou,et al.  Simple Traffic Modeling Framework for Machine Type Communication , 2013, ISWCS.

[13]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[14]  Bojan Mrazovac,et al.  Performance evaluation of using Protocol Buffers in the Internet of Things communication , 2016, 2016 International Conference on Smart Systems and Technologies (SST).

[15]  Emad H. Al-Hemiary,et al.  Performance Analysis of Internet of Things Protocols Based Fog/Cloud over High Traffic , 2018 .