Graph Oriented Databases and Suitability for the Internet of Things

With the growing digitalization of the industry, more data is available and can be used to improve production processes. The amount of data created depends on the individual use case but still, it needs to be stored to be useful. Since there are multiple databases available it can be difficult to choose the right one for an individual scenario. Current graph database benchmarks only cover social network graphs, which differ from the data structure present in the industry. We have chosen the databases to use for our testing from other studies covering benchmarking graph databases to be able to compare the results and look at similarities in behavior. To evaluate different databases, we first looked up existing benchmarks and choose the best one for our research. In the benchmarking program we looked at the creation of data and how it can be stored and retrieved. The same exact dataset should be used for all databases equally to eliminate the variation that comes with generating data during each benchmark run. Workloads has been designed to investigate the performance of graph databases with industrial data and the production environment has been simulated. With the databases and benchmark set up, we executed the workloads and evaluated the results to conclude whether current databases are suitable for the industrial internet of things .