And QUIC meets IoT: performance assessment of MQTT over QUIC F´ atima Fern´ andez † , Mihail Zverev † , Pablo Garrido † , Jos´ e R. Ju´ arez † , Josu Bilbao † , Ram´ on Ag¨ uero ¶ † Ikerlan Technology Research Centre, Basque Research Technology Alliance (BRTA) Arrasate/Mondragon, Spain {ffernandez, mzverev, pgarrido, jrjuarez, jbilbao}@ikerlan.es ¶ Dpt. of Communication Engineering University of Cantabria Santander, Spain ramon@tlmat.unican.es Abstract—We study the performance of the Message Queuing Telemetry Transport Protocol (MQTT) over QUIC. QUIC has been recently proposed as a new transport protocol, and it is gaining relevance at a very fast pace, favored by the support of key players, such as Google. It overcomes some of the limitations of the more widespread alternative, TCP, especially regarding the overhead of connection establishment. However, its use for Internet of Things (IoT) scenarios is still under consideration. In this paper we integrate a GO-based implementation of the QUIC protocol with MQTT, and we compare the performance of this combination with that exhibited by the more traditional MQTT/TLS/TCP approach. We use Linux Containers and we emulate various wireless network technologies by means of the ns- 3 simulator. The results of an extensive measurement campaign, show that QUIC protocol can indeed yield good performances for typical IoT use cases. Index Terms—Quic UDP Internet Connections (QUIC), In- ternet of Things (IoT), Message Queuing Telemetry Transport (MQTT), Lossy Networks, Performance Analysis I. I NTRODUCTION Industry and companies are witnessing a deep transforma- tion, known as Industry 4.0, which is indeed considered as the fourth industrial revolution. This term, introduced in 2011 by the German government, sprung as a novel initiative towards the digitalization of its industry, based on four principles: inter- connection, information transparency, decentralized decisions and technical assistance [1]. Industry 4.0 yields a remarkable productivity boost, thanks to (among other functionalities) the collection and analysis of real-time data [2]. One of the key technologies behind this industry digital transformation is the IoT paradigm, which enables the con- nection of industrial components, to collect data, monitor sys- tems, exchange information, and analyze the environment [2]. However, the development of industrial 4.0 applications and services introduces stringent requirements. It is worth men- tioning the following ones: low latency communications, high reliability and availability, energy and cost efficiency, and security and privacy [3]. Energy and cost efficiency is a pivotal feature. In order to monitor the industrial environment, a huge deployment of sensors and connected devices is required, but this deployment should not cause a cost increase of the manufacturing pro- cess. Besides, the connected devices and sensors must reach different locations and isolated areas and, in many cases, this will imply the use of battery-powered elements, which should stay alive for a long time. Low latency communications are also essential in many industrial applications. Very fast responses might be crucial in some scenarios in order to enhance manufacturing processes, to guarantee the security of many involved elements, and to provide the chance to enable remote control functions in real industrial scenarios. The well-known MQTT protocol [4], [5] is widely used in IoT applications, due to its small code footprint, easy integration and good performance. It traditionally lays, as its transport protocol, on TCP [6], which offers a reliable end- to-end communication service. However, TCP suffers from many disadvantages, specially those related to the ossification of internet protocols, which are not able to adapt to the fast- changing technologies [7]. Dealing with lossy networks has been shown to be very challenging for TCP, drastically jeop- ardizing its performance and increasing end-to-end delay [8]. Moreover, the extremely low communication delay require- ment of some applications is not appropriately guaranteed by TCP [9]. Therefore, many alternatives have emerged, some of them being TCP modifications or updates, such as SCTP [10], Real-Time TCP [11] and Network Coded TCP [12] among others. One of these alternatives is QUIC [13], an experimental transport protocol, whose main design principles were to reduce connection establishment and transport latencies, as well as to improve security standards with default end-to-end encryption in HTTP-based applications [14]. In this paper we propose to integrate IoT protocols (MQTT in particular) with QUIC, in order to assess the performance of such combination and to analyze the benefits that novel transport protocol approaches can bring to the IoT realm, compared to legacy ones. The contributions of this paper are: • Integration of a fully operational MQTT implementation (in GO programming language) with a QUIC GO imple- mentation.