Fog Assisted Cloud Models for Smart Grid Architectures- Comparison Study and Optimal Deployment
share
Cloud Computing (CC) serves to be a key driver for fulfilling the store and compute requirements of a modern Smart Grid (SG). However, since the datacenters are deployed in concentrated and far remote areas, it fails to guarantee the quality of experience (QoE) attributes for the SG services, viz. latency, bandwidth, energy consumption, and network cost. Fog Computing (FC) extends the processing capabilities into the edge of the network, offering location-awareness, low latency, and latency-sensitive analytics for mission critical requirements of SG. In this work, we first examine the current state of cloud based SG architectures and highlight the motivation(s) for adopting FC as technology enabler for sustainable and real-time SG analytics. Then we present a hierarchical FC architecture for supporting integration of massive number of IoT devices into future SG. Following this architecture we proposed a cost optimization framework that jointly investigates data consumer association, workload distribution, virtual machine placement and QoS constraints towards viable deployment of FC model over SG networks. The formulated MINLP problem is then solved using Modified Differential Evolution (MDE) algorithm. Comprehensive evaluation of proposed framework on real world parameters shows that for an infrastructure with nearly 50 requesting real-time services, the overall service latency for fog computing get reduced to almost half of that of generic cloud paradigm. It is also observed that the fog assisted cloud framework lowers the aggregated electricity consumption of the pure cloud computing paradigm by more than 40
READ FULL TEXT