With the recent development in the Internet of Things (IoT), big data, and machine learning, the number of services has dramatically increased. These services are heterogeneous in terms of the amount of resources and quality of service (QoS) requirements. To cope with the limitations of Cloud infrastructure providers (CIPs) for latency-sensitive services, many Fog infrastructure providers (FIPs) have recently emerged and their numbers are increasing continually. Due to difficulties such as the different requirements of services, location of end-users, and profile cost of IPs, distributing services across multiple FIPs and CIPs has become a fundamental challenge. Motivated by this, a flexible and scalable platform, FLEX, is proposed in this work for the service placement problem (SPP) in multi-Fog and multi-Cloud computing. For each service, FLEX broadcasts the service's requirements to the resource managers (RMs) of all providers and then based on the RMs' responses, it selects the most suitable provider for that service. The proposed platform is flexible and scalable as it leaves it up to the RMs to have their own policy for service placement. The problem is formulated as an optimization problem and an efficient heuristic algorithm is proposed to solve it. Our simulation results show that the proposed algorithm can meet the requirements of services.

No abstract

Provisioning services for Internet of Things (IoT) devices leads to severalchallenges: heterogeneity of IoT devices, varying Quality of Services (QoS)requirements, and increasing availability of both Cloud and Fog resources. Thelast of these is most significant to cope with the limitations of Cloud infrastructureproviders (CIPs) for latency-sensitive services. Many Fog infrastructure providers(FIPs) have recently emerged and their number is increasing continually. FLEX isproposed in this work as a platform for selecting a location for service placementin a multi-Fog and multi-Cloud environment. For each service, FLEX broadcastsservice requirements to the resource managers (RMs) of the available Fog andCloud service providers and then selects the most suitable provider for that service.FLEX is scalable and flexible as it leaves it up to the RMs to have their ownpolicy for the placement of submitted services. Service placement and resource selectionhas been formulated as an optimization problem and an efficient heuristicalgorithm is proposed to solve it. Results show that the proposed algorithm canbe used across both Cloud and Fog-based providers.