The authors have expertise with Oracle databases and use examples from the enterprise products this software vendor provides to make their point regarding security of highly distributed networks. One of the more valuable aspects of this specific paper is the focus on how to create a multilevel secure environment in an enterprise. The authors have done enterprise-level database security work in their careers and this article and research communicate their expertise clearly. In the article Data Security: A Security Implementation for Relational Database Management Systems (Nilakanta, 1989) the author contends that information architectures must rely on a stable database management system (DBMS) to scale securely and reliably across an enterprise. The author provides insights into several different security procedures and approaches to defining a secured operating environment for enterprise-wide DBMS implementations and use. There are also guidelines for defining security clearances and recommendations on hwo best to use encryptions for backing up stored, confidential data. The author also contends that given the increasingly easy-to-use interfaces of relational database systems in the enterprise there is a corresponding need for more robust, thorough group- and rules-based applications for better managing this and future generations of databases deliberately designed for widespread adoption and use. The author also provides insights into how the central Database Administrator needs to also define object-based rights and permissions across the entire complex of databases in an enterprise, including the defining of access rights by role, group and workflow that a given employee is involved with.

In the article Multilevel Security Issues in Distributed Database Management Systems II (Thuraisingham, 1991) the author...

He further shows how the use of multi-level-based security taxonomy can significantly improve overall performance of heterogeneous databases architectures, incouding those based on the multilevel secure (MLS) model. The author recommends the development and continued fine-tuning of a two-phase commitment protocol that is synchronous in operation, capable of suppoirting a multitier approach to database security model development and role, group and workflow-based security protocol definition. The author provides insights into how time-stamping techniques can also be invaluable in defining how a multilayer-based strategy for managing enterprise security can be used effectively. By combining the MLS-based taxonomies of database design with the DDBMS architectural structures and constraints, the author provides an overview of how to create an effective, scalable multilevel security model for distributed databases in global enterprises.