Survey on Service Migration, load optimization and Load Balancing in Fog Computing Environment

The Internet of Things (IoT) devices are characterized by the less storage capacity and limited computational power. This calls for a great amount of cloud services to provide the required support to these smart devices. However such cloud services performance is restrained by the performance offered by the respective servers in Data Centers, which host the cloud services. Such Data Centers are usually located at a nearby proximity to the core networks, which are at large distances from the edge or field locations. This results in end to end higher latency. This problem is further compounded by the big amount of data that comes from IoT devices, which are required to be examined in real-time. However, in real-life cases such as health condition monitoring, urgently responsive and other latency sensitive applications, the latency caused by the transfer of the data to cloud server and getting it analyzed to reach conclusions, which have to be sent back to the edge is very large and therefore not realistic or acceptable. So, a new paradigm known as Fog Computing is emerging which seamlessly integrates edge devices on one hand with cloud resources on the other. In large traffic situations load balancing will become essential. Application Migration across Fog Servers may also become warranted, either for load balancing or due to physical movement of IoT devices. In this paper we provide a survey of techniques of service migration, load optimization and load balancing in the Fog eco system.