Chapter 5. Security Architecture and Design

This chapter presents the following:

• Computer hardware architecture

• Operating system architectures

• Trusted computing base and security mechanisms

• Protection mechanisms within an operating system

• Various security models

• Assurance evaluation criteria and ratings

• Certification and accreditation processes

• Attack types

Computer and information security covers many areas within an enterprise. Each area has security vulnerabilities and, hopefully, some corresponding countermeasures that raise the security level and provide better protection. Not understanding the different areas and security levels of network devices, operating systems, hardware, protocols, and applications can cause security vulnerabilities that can affect the environment as a whole.

Two fundamental concepts in computer and information security are the security policy and security model. A security policy is a statement that outlines how entities access each other, what operations different entities can carry out, what level of protection is required for a system or software product, and what actions should be taken when these requirements are not met. The policy outlines the expectations that the hardware and software must meet to be considered in compliance. A security model outlines the requirements necessary to properly support and implement a certain security policy. If a security policy dictates that all users must be identified, authenticated, and authorized before accessing network resources, the security model might lay out an access control matrix that should be constructed so it fulfills the requirements of the security policy. If a security policy states that no one from a lower security level should be able to view or modify information at a higher security level, the supporting security model will outline the necessary logic and rules that need to be implemented to ensure that under no circumstances can a lower-level subject access a higher-level object in an unauthorized manner. A security model provides a deeper explanation of how a computer operating system should be developed to properly support a specific security policy.

Note

Individual systems and devices can have their own security policies. These are not the organizational security policies that contain management’s directives. The systems’ security policies, and the models they use, should enforce the higher-level organizational security policy that is in place. A system policy dictates the level of security that should be provided by the individual device or operating system.

Computer security can be a slippery term because it means different things to different people. Many aspects of a system can be secured, and security can happen at various levels and to varying degrees. As stated in previous chapters, information security consists of the following main attributes:

• Availability Prevention of loss of, or loss of access to, data and resources

• Integrity Prevention of unauthorized modification of data and resources

• Confidentiality Prevention of unauthorized disclosure of data and resources

These main attributes branch off into more granular security attributes, such as authenticity, accountability, nonrepudiation, and dependability. How does a company know which of these it needs, to what degree they are needed, and whether the operating systems and applications they use actually provide these features and protection? These questions get much more complex as one looks deeper into the questions and products themselves. Companies are not just concerned about e-mail messages being encrypted as they pass through the Internet. They are also concerned about the confidential data stored in their databases, the security of their web farms that are connected directly to the Internet, the integrity of data-entry values going into applications that process business-oriented information, internal users sharing trade secrets, external attackers bringing down servers and affecting productivity, viruses spreading, the internal consistency of data warehouses, and much more.

These issues not only affect productivity and profitability, but also raise legal and liability issues with regard to securing data. Companies, and the management that runs them, can be held accountable if any one of the many issues previously mentioned goes wrong. So it is, or at least it should be, very important for companies to know what security they need and how to be properly assured that the protection is actually being provided by the products they purchase.

Many of these security issues must be thought through before and during the design and architectural phase for a product. Security is best if it is designed and built into the foundation of operating systems and applications and not added as an afterthought. Once security is integrated as an important part of the design, it has to be engineered, implemented, tested, audited, evaluated, certified, and accredited. The security that a product provides must be rated on the availability, integrity, and confidentiality it claims to provide. Consumers then use these ratings to determine if specific products provide the level of security they require. This is a long road, with many entities involved with different responsibilities.

This chapter takes you from the steps that are necessary before actually developing an operating system to how these systems are evaluated and rated by governments and other agencies, and what these ratings actually mean. However, before we dive into these concepts, it is important to understand how the basic elements of a computer system work. These elements are the pieces that make up any computer’s architecture.