In order to meet the diverse requirements for integrity protection, it is necessary to go beyond isolated systems and multi-level access authorizations. If legal requirements even stipulate forensic proof of data integrity, far-reaching measures must be taken to protect sensitive company data from destruction, manipulation or criminally motivated third-party encryption. This applies to the central data sets in which the company stores its trade secrets as well as to the digital infrastructure on which this data is processed.

A data-centric approach is formulated for a holistic view of requirements.

Data-centric approach

The data-centric approach is used here, since this allows the clearest formulation of the structured descriptions required to protect the information-bearing objects. For all considerations, the focus in each case is on the data to be protected. This applies not only to trade secrets such as design plans or descriptions, calculations, contracts or configuration data, but generally to all digital assets that are valuable to the company.

When formulating rules for processing sensitive data, all processing options, such as opening and modifying files, saving, copying, renaming, moving, deleting, etc., are taken into account. Furthermore, the person or group authorized to do so, the time, but also the application used for this purpose are also taken into account.

In the example shown here, which is intended to illustrate the methodology of the requirements representation and analysis, the Use Case deals with the permitted as well as the unpermitted manipulation of data. Here, data generally stands for a single file, several files, a complete directory or a directory tree, but also for a complete storage system.

For the visual representation of the Use Case, all possible users and user groups are positioned concentrically around the data positioned in the center. For this simple example, no detailed examination of the data through data classification, context data or an analysis of the file content is described. Also not in focus for this use case is the number of files processed or the volume of data within a given time period.

Implemented here is a simple distinction for permissible and impermissible file operations by different users. A more differentiated formulation can also describe applications and processes.

Even without enforcing access restrictions, an audit trail can be created for all file manipulations. Access violations or unauthorized data manipulations that constitute a rule violation are detected, reported and prevented instantaneously. It is irrelevant whether this attack is carried out by a hacker directly or by a computer infiltrated with malicious code that wants to access the protected data.

Even if a computer was previously infected with malicious code by another system, this does not pose a relevant threat to the confidential data.

The enormous advantage for a data-centric approach at this point is the possibility of applying the superposition principle. Multiple, individually formulated use cases can be superimposed. Individual rules can be formulated in isolation to fit into a larger and more complex whole.

This superposition principle, known from physics, is used for linear problems in many areas (forces in classical mechanics, optics, but also states in quantum mechanics). The framework used also avoids overlapping or contradictory rules.