The concept of polyinstantiation is not well understood in the academia as many scholars struggle to explain its origin and how it occurs. From various studies, it is claimed that data integrity and polyinstantiation are inseparable. The use of DBMS and MLS requires the application of a Mandatory Access Control Code to authenticate unauthorized users. Due to this, polyinstantiation is often related to multi-level data systems. The occurrence is related to specific data and not to particular technologies. Similarly, it occurs in tuples and not when data area allocated to relations (Jajodia & Sandhu, 1994).
Several types of polyinstantiation exist. In Entity Polyinstantiation, similar primary key identifiers are given to more than one tuples in a multi-level relation. This is where a low level user accords an entity an identifier that is already in use by a high level user for another entity. In attribute polyinstantiation, entities with the same attributes are given different identifiers by the low level and high level users. Cover stories on the other hand are a type of polyinstantiation which occurs when entities are given attributes that are lower than the actual attributes (Lunt, 1991).
Unnecessary polyinstantiation occurs when an entity key is associated with a real world entity by the high level users in case where the low level attributes are already in existence. Polyinstantiation can be either visible where the high level users try to enter data that is already at the lower level or invisible. In invisible polyinstantiation, the low level users add data to already existing high level entities. In each case, the effort is met with option to update information or to decline the polyinstantiation in the case of visible and invisible polyinstantiation respectively. Accepting the changes and allowing polyinstantiation can result in the formation of a cascade of tuples leading to the generation of multiple cover stories.
Polyinstantiation can occur when a user adds a cover story to high level information deliberately; when a user fails to check whether there has been referencing at a lower level when creating relations and when other tuples are created without deletion of the previously existing tuples. Polyinstantiation can result in several problems for the database designer, user and the administrator. For instance, it brings about a conflict between the database integrity and information confidentiality. In addition to this, polyinstantiation can also result in real world integrity interference through introduction of ambiguity. There is therefore need to prevent polyinstantiation in databases. This can be achieved through control of access of low level users to the views of the higher level users in multi-level system. This is because polyinstantiation results in reduced secrecy in information use.
In order to effectively address the problems caused by polyinstantiation, Qian (1994) recommends the dissociation of confidentiality and data integrity. Through inference, elimination of interference from low datum in multi level systems would not occur. Alternatively, polyinstantiation may be taken as a usual occurrence in multi level systems. This may however pose two challenges i.e. managing tuple propagation and ensuring identities relate to real life entities. The second challenge can be tackled through the use of the modification approach or the sea view approach.
On the other hand, the first challenge can only be addressed through the augmenting the identifier with the security access controls. Other control methods may be through the belief approach or the derived value approach. The belief approach involves the addition of a replica between the tuple level and the element instantiation level (Smith & Winslet, 1992). It is thus recommended that development of DBMS should consider the concept of polyinstantiation and effectively address it at all levels.
Jajodia, S., & Sandhu, R. (1994). Solutions to the polyinstantiation problem. GEORGE MASON UNIV FAIRFAX VA.
Qian, X. (1994, May). Inference channel-free integrity constraints in multilevel relational databases. In Research in Security and Privacy, 1994. Proceedings., 1994 IEEE Computer Society Symposium on (pp. 158-167). IEEE.
Qian, X. (1992). Integrity, Secrecy, and Inference Channels. Proceedings of the 5th Rome Lab Database Symposium.Smith, K., & Winslett, M. (1992, August). Entity modeling in the MLS relational model. In VLDB (pp. 199-210).
Lunt, T. (1991, June). A new polyinstantiation integrity constraint for multilevel relations. In Computer Security Foundations Workshop III, 1990. Proceedings (pp. 159-165). IEEE.
Would you like us to custom write a computer science research paper like the one summarized above on polyinstantiation? One thing we would want you to note is that we have some of the best academic writers you will find anywhere. Our writers are some of the highest rated in academic writing. You can therefore trust us to write high quality research papers for you and make sure they are submitted to you on time. To read some more samples of research papers kindly browse through our blog.