Information Systems (IS) are in constant evolution to cope with new technologies like cloud computing and Big data. Nevertheless, these new technologies bring several security issues, some caused by the numerous access points which make companies prone to an increasing number of attacks. The architecture of an information system can be structured in three main layers: data layer, application layer, and technology layer. Security must be reinforced in each of the previous layers to prevent internal and external attacks. In this paper, we focus on the security of the data layer, thus we survey the research work dealing with security related to this layer. For that, we propose a data meta-model with a high level of abstraction that shows the relationship between data, security issues, and security mechanisms covering the three aspects of security (confidentiality, integrity, availability) and considering different types of data and novel technologies (big data and cloud computing). We use the main concepts of this meta-model to classify security mechanisms according to security goals, to compare the different approaches proposed in the literature, and to identify open challenges in the data security field. Our data meta-model can be easily extended and serves as a reference to compare other approaches in the literature.