Latency Critical Data Processing in Cloud for Smart Grid Applications

Abstract

For the harmonious operation among various entities, smart grid requires sharing and processing a huge amount of data and information. In the regard, cloud computing can provide the leverage of storage, processing and management of data using a shared pool of configurable resources over the Internet. In some cases of control and monitoring, low latency data processing is mandatory. To address this, we propose two types of priority data processing in the cloud for the smart grid applications. In the first type of data processing called preemptive priority, the priority data is processed ceasing the processing of an on-going general packet. On the other hand, in the second type called as non-preemptive priority, the priority packet is processed after the execution of an ongoing general packet. The general packets are processed on first come first served. In this paper, we evaluate the performance of the two data processing methods in the scenario of a smart grid. Based on the results, preemptive data processing is recommended for extreme latency critical data while non-preemptive is suitable for latency critical data.