Received February 19, 2017, accepted March 9, 2017, date of publication March 15, 2017, date of current version April 24, 2017. Digital Object Identifier 10.1109/ACCESS.2017.2682499 Delay Mitigation in Offloaded Cloud Controllers in Industrial IoT SAAD MUBEEN 1 , (Senior Member, IEEE), PAVLOS NIKOLAIDIS 1 , ALMA DIDIC 1 , HONGYU PEI-BREIVOLD 2 , KRISTIAN SANDSTRÖM 3 , AND MORIS BEHNAM 1 , (Member, IEEE) 1 Mälardalen University, 721 23 Västerås, Sweden 2 ABB Corporate Research, 722 26, Västerås, Sweden 3 Swedish Institute of Computer Science, 722 13, Västerås, Sweden Corresponding author: S. Mubeen (saad.mubeen@mdh.se) The work was supported in part by the Swedish Foundation for Strategic Research through the Project Future factories in the Cloud, in part by the Swedish Governmental Funding from Strategic Research Area through the project XPRES, and in part by the Swedish Knowledge Foundation through the Project Preview. ABSTRACT This paper investigates the interplay of cloud computing, fog computing, and Internet of Things (IoT) in control applications targeting the automation industry. In this context, a prototype is developed to explore the use of IoT devices that communicate with a cloud-based controller, i.e., the controller is offloaded to cloud or fog. Several experiments are performed to investigate the consequences of having a cloud server between the end device and the controller. The experiments are performed while considering arbitrary jitter and delays, i.e., they can be smaller than, equal to, or greater than the sampling period. This paper also applies mitigation mechanisms to deal with the delays and jitter that are caused by the networks when the controller is offloaded to the fog or cloud. INDEX TERMS Industrial IoT, fog computing, cloud computing, industrial automation systems. I. INTRODUCTION Cloud computing [1] and Internet of Things (IoT) [2], [3] are two notable concepts that have evolved significantly over the past few years. Cloud computing is an operational scheme that provides network-based services such as computational power, storage and networking to users within many indus- trial and application domains. It offers a pool of virtualized computing resources at various levels, covering infrastruc- ture, platforms or software delivered to users as on-demand services from the cloud. In this way, cloud computing is changing the services consumption and delivery platform as well as the way businesses and users interact with IT resources. IoT extends the cloud computing concept beyond com- puting and communication to include everything, i.e., also the physical devices. Industrial IoT uses sensors, machine- to-machine (M2M) collaboration and various technologies to gather and analyze data from physical and virtual world for optimized operations and providing services. In IoT, devices are connected through a network. They share data, informa- tion and even resources to accomplish their goal or increase total system intelligence. Accordingly, cloud computing and IoT can provide services to consumers and businesses, allow- ing organizations to become more agile and flexible in pur- suing new revenue streams and new business models. Despite numerous advantages of cloud computing, there are some limitations such as high delays that render cloud computing unfavorable to the industrial control systems that have low-delay requirements. In order to overcome the drawbacks of cloud computing, a local cloud computing architecture called the fog computing has been introduced recently [4]. According to Cisco, fog computing extends the cloud computing away from the cloud computing data centers and towards the edge of the network [5]. A representation of the three-tier fog computing architec- ture proposed by Cisco is shown in Fig. 1. The first tier of the architecture gets data from the local embedded devices and sensors. This tier is used for the M2M interactions and sup- ports real-time systems with stringent timing requirements. A real-time system is required to provide its logically correct response within the time that is mandated by the specified timing requirement(s). The M2M interaction is the key aspect to increase the intelligence of the ‘‘things’’ [6], [7]. The response times provided by the first tier are in the order of 4418 2169-3536  2017 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. VOLUME 5, 2017