Towards Modeling Large-Scale Data Flows in a Multidatacenter Computing System With Petri Net

There are several use cases that involve the need to transfer data between datacenters when processing large-scale data sets. Data transmission in a multidatacenter computing system has new characteristics that are hard to express and handle using traditional methods. Thus, it is necessary for a new transmission strategy to be developed for users of multidatacenter computing systems. This paper gives a general data model, a network model, and a multidatacenter model to describe the research problem clearly. We propose four new methods to meet the new features of data transmission among datacenters. Based on these models and methods, we propose a new data flow transmission model and analyze its validity.