Security Architecture and
Design: Part II
Chao-Hsien Chu, Ph.D.
College of Information Sciences and Technology
The Pennsylvania State University
University Park, PA 16802
chu@ist.psu.edu
IST 515
The Transformation Process
The security policy provides the abstract goals and the security
model provides the do’s and don’ts necessary to fulfill the goals
Security
Policy
• Abstract objectives,
goals and
requirements
• Rules or practice
• Framework
Security
Model
• Mathematical
relationship and
formulas
• Specifications
• Data structure
Programming
• Computer code
• GUI – check box
• Product or
• System
Security Policy
• Outlines the security requirements for an organization.
• Is an abstract term that represents the objectives and goals a
system must meet and accomplish to be deemed secure and
acceptable.
• Is a set of rules and practices that dictates how sensitive
information and resources are managed, protected, and
distributed.
• Expresses what the security level should be by setting the
goals of what the security mechanisms are supposed to
accomplish.
• Provides the framework for the systems’ security architecture.
Security Model
• Is a symbolic representation of a policy, which
– Outlines the requirements needed to support the security
policy and how authorization is enforced.
– Maps the abstract goals of the policy to information
system terms by specifying explicit data structures and
techniques that are necessary to enforce the security
policy.
– Maps the desires of the policymakers into a set of rules
that computer system must follow.
• Is usually represented in mathematics and analytical ideas,
which are then mapped to system specifications, and then
developed by programmers through programming codes.
Security Policy
“Subjects need to be authorized to access objects.”
Example
Security Model
• Derive the mathematical
relationships and formulas
explaining how x can access y
only through outlined specific
methods.
• Develop specifications to provide
a bridge to what this means in a
computing environment and how
it maps to components and
mechanisms that need to be coded
and developed.
• Write the program code to
produce the mechanisms that
provide a way for a system to use
access control lists and give
administrators some degree of
control. This mechanism presents
the network administrator with a
GUI representation, like check
boxes, to choose which subjects
can access what objects, within
the operating system.
Security Models
Security Requirements
Security Models
• Bell-LaPadula Model (1973)
• Biba Model (1977)
• Confidentiality
• Clark-Wilson Model (1987)
• Integrity
• Access Control Matrix
• Availability
• Information Flow Model
• Noninterference Model
• Chinese Wall Model
• Lattice Model
Bell-LaPadula Model
• Funded by the U.S. government, Bell-LaPadula
model is the first mathematical model of a
multilevel security policy. Because users with
different clearances use the system, and the system
processes data with different classifications.
• Is a state machine model that enforce the
confidentiality aspects of access control, but not
with integrity or availability
• Is an information flow security model as it ensures
information does not flow in an insecure manner.
• All mandatory access control (MAC) model are
based on the Bell-LaPadula model.
Bell-LaPadula Model Properties
• The Simple Security Property (ss Property) states that a
subject at a given security level cannot read data that
resides at a higher security level (No Read Up).
• The * (star) Security Property states that a subject in a
given security level cannot write information to a lower
security level. (No Write Down).
• The Strong Star Property states that a subject that has
read and write capabilities can only perform those
functions at the same security level, nothing higher and
nothing lower. A subject to be able to read and write to
an object, the clearance and classification must be equal.
The Bell-LaPadula Model
Layer of
Higher Secrecy
Χ
Reading
Secrets
Read
Layer of
Lower Secrecy
Χ
Write
Χ
Simple
Security
Property
Divulging
Secrets
Star (*)
Property
Reading
Secrets
Read/Write
Χ
Divulging
Secrets
Strong
Star (*)
Property
Biba Security Model
• Developed in 1977, the Biba integrity model
mathematically describes read and write
restrictions based on integrity access classes of
subjects and objects. It is the first model to
address integrity.
• Is an information flow model as it is concerned
about data flowing from one level to another.
• The model looks similar to the Bell-LaPadula
Model; however, the read-write conditions are
reversed.
Biba Integrity Model Axiom
• The Simple Integrity Axiom: States that a subject at
one level of integrity is not permitted to observe
(read) an object of a lower integrity. No Read Down.
• The * (Star) Integrity Axiom: States that an object at
one level of integrity is not permitted to modify
(write to) an object of a higher level of integrity. No
Write Up.
• Invocation property states that a subject at one level
of integrity cannot invoke (call up) a subject at a
higher level of integrity.
The Biba Model
Layer of
Higher Secrecy
Χ
Read
Layer of
Lower Secrecy
Χ
Contamination
Write
Get
Contaminated
Simple
Integrity
Property
Integrity
Star (*)
Property
The Invocation Property
• The Biba model can be extended to include an
access operation called invoke. A subject can
invoke another subject, such as a software utility,
to access an object.
• The subject cannot send message (logical request
for service) to subjects of higher integrity.
Subjects are only allowed to invoke utilities or
tools at the same or lower integrity level
(otherwise, a dirty subject could use a clean tool
to access or contaminate a clean object).
Clark-Wilson Integrity Model
The model uses the following elements:
• Users: Active agents.
• Transformation Procedures (TPs): Programed abstract
operations, such as read, write and modify.
• Constrained Data Item (CDI): A data item whose integrity
is to be preserved. Can only be manipulated by TPs.
• Unconstrained Data Item (UDI): Data items outside of the
control area of the modeled environment such as input
information. Can be manipulated by users via primitive
read and write operations.
• Integrity Verification Procedure (IVP): Check the
consistency of CDIs with external reality.
Elements of Clark-Wilson Model
Users
UDI
CDI
CDI 1
CDI 2
TP
IVP
CDI 3
Log CDI
Clark-Wilson Model
The three main rules of integrity models:
• Prevent unauthorized users from making
modifications
• Prevent authorized users from making improper
modifications (separation of duties)
• Maintain internal/external consistency (wellformed transaction)
Clark-Wilson model addresses each of these goals.
Biba model only addresses the first goal.
Clark-Wilson Model
• Developed by Clark-Wilson in 1987, the model
addresses the integrity requirements of
applications.
• Clark-Wilson model enforces the three goals of
integrity by using access triple (subject,
software TP, and object), separation of duties,
and auditing. It enforces integrity by using wellformed transactions (through access triple) and
separation of user duties.
Information Flow Model
• Information flow model deals with any kind of information
flow, which help architects and developers make sure their
software does not allow information to flow in a way that
can put the system or data in danger.
• One way that the information flow model provides
protection is by ensuring that covert channels do not exist
in the code.
• The Bell-LaPadula model focuses on preventing
information from flowing from a high security level to a
low security level.
• The Biba model focuses on preventing information from
flowing from a low security level to a high security level.
Other Models
• State machine model is an abstract mathematic model that
uses state variables to represent the system state. Failure of a
state machine should fail in a secure state.
• Non-interference model ensures that actions at a higher level
(domain) cannot interfere with actions at a lower level.
• Graham-Denning Modem defines a set of eight primitive
protection rights in terms of commands that a specific
subject can execute on an object.
• Brewer and Nash Model (Chinese Wall Model) allows for
dynamically changing access controls to protect against
conflicts of interest.
Security Protection Mechanisms
• Protection mechanisms are used to ensure the
separation between objects in a system.
• Active protection mechanism. It prevents access
to an object if the access is not authorized.
• Passive protection mechanism. It prevents or
detects unauthorized use of the information
associated with an object, even if access to the
object itself is not prevented. In most cases, these
techniques use cryptographic techniques to
prevent unauthorized disclosure of information or
checksum techniques to detect an unauthorized
alteration of an object.
Protection Mechanisms 1
• Trusted Computing Base (TCB)
• Trusted Computer System
• Abstraction, Encapsulation, and
Information Hiding
• Security Perimeter
• Trusted Path
• Labeling and Classification
Protection Mechanisms 2
• Centralized backup for desktop systems
• Control of software on desktop systems
• Encryption / File encryption
• Appropriate access controls
• Robust access control and biometrics
• Protection of applications and database
• Protection domain, disks, systems, laptops
Protection Mechanisms 3
• Separation of privileged process and others
• Logging of transaction and transmission
• Email and download/upload policies
• Security awareness and regular training
• Graphical user interface mechanism
• Security formal methods in software development,
change control, configuration management, and
environmental change
• Disaster recovery and business continuity planning, for
all systems including desktop, file system and storages,
database and applications, data and information
Factors for Security Product Selection
•
•
•
•
•
•
Security.
Cost.
Flexibility.
Environmental.
User interface.
System
administration.
• Future development
of a product.
• Process.
• Functionality.
• Effectiveness.
• Assurance.
Security / System Evaluation
• A security evaluation examines the security-relevant
parts of a system including:
– Trusted computing base (TCB)
– Access control mechanisms.
– Reference monitor
– Kernel
– Protection mechanisms.
• There are different methods of evaluating and assigning
assurance levels to systems, as various parts of the
world look at computer security differently and rate
some aspects of security differently.
Security Evaluation Standards
• Trusted Computer Security Evaluation Criteria (TCSEC):
– 1985 by the National Computer Security Center (NCSC)
– Also known as Orange Book of the Rainbow Series.
– Address Confidentiality.
• The Trusted Network Interpretation (TNI):
– 1987, known as Red Book
– Address network and telecommunications
• Information Technology Security Evaluation Criteria (ITSEC):
– Drafted in 1990 and endorsed by the Council of the European
Union in 1995.
– Include integrity and availability as well as confidentiality as
security goals.
• The Common Criteria (CC):
– Based on the U.S. Federal Criteria that expanded on the ITSEC.
– An international standard to evaluate trust.
Orange Book - TCSEC
• The U.S. Department of Defense (DoD) developed the Trusted
Computer System Evaluation Criteria (TCSEC) to evaluate
operating systems, applications, and different products.
• These evaluation criteria are published in a book with an
orange cover, which is called, appropriately, the Orange Book.
• The Orange Book is used to review the functionality,
effectiveness, and assurance of a product during its evaluation,
which can be used to evaluate whether a product contains the
security properties the vendor claims it does and whether the
product is appropriate for a specific application or function.
Orange Book - TCSEC
Topics
• Security policy.
• Labels. Marking of objects.
• Identification of subjects.
• Accountability.
• Life-cycle assurance.
• Documentation.
• Continuous protection.
Classification Systems
A: Verified Protection.
B: Mandatory Protection:
B1: Labeled security.
B2: Structured protection.
B3: Security domains
C: Discretionary Protection:
C1: Discretionary security
protection
C2: Controlled asset protection
D: Minimal Protection
Rainbow Series
• Security practitioners have pointed out the deficiencies in
the Orange Book:
– It looks specifically at the operating system and not at
other issues like networking and database.
– It focuses mainly on one attribute of security –
confidentiality, and not on integrity and availability.
– It works with government classification and not the
protection classifications commercial industries use.
– It has a relatively small number of rating. Many different
aspects of security are not evaluated and rated.
• More books were written to extend the coverage of the
Orange Book, which are collectively called Rainbow Series.
The Red Book - TNI
• The Trusted Network Interpretation (TNI), also called the Red
Book, addresses security evaluation topics for networks and
network components, including local area networks and wide
area internetwork systems.
• Like Orange Book, the Red Book only provides a framework
for securing different types of networks, and does not supply
specific details on how to implement security mechanisms.
• The Red Book rates confidentiality of data and operations that
happen within a network and the network products. Data and
labels need to be protected from unauthorized modification,
and the integrity of information needs to be ensured. The
source and destination mechanisms used for messages are
evaluated and tested to ensure modification is not allowed.
Common Criteria
• The Orange Book and the Rainbow Series provide evaluation
schemes that are too rigid for the business world. ITSEC
attempted to provide a more flexible approach by separating
the functionality and assurance. However, this add much
complexity and results in too many classifications to keep
straight.
• The International Organization for Standardization (ISO)
identified the need of international standard evaluation criteria
in security, which resulted in the development of Common
Criteria. The Common Criteria was developed through a
collaboration among national security standards organizations
within the United States, Canada, France, Germany, the United
Kingdom, and the Netherlands.
Common Criteria Assurance Levels
• Under the Common Criteria model, an evaluation is carried
out on a product and it is assigned an Evaluation Assurance
Level (EAL).
• The Common Criteria has seven assurance levels:
– EAL1: Functionally tested
– EAL2: Structurally tested
– EAL3: Methodically tested and checked
– EAL4: Methodically designed, tested and reviewed
– EAL5: Semiformally designed and tested
– EAL6: Semiformally verified design and tested
– EAL7: Formally verified design and tested.
Certification
• Certification is the comprehensive evaluation of the
technical and non-technical security features of an
information system and the other safeguards, which are
created in support of the accreditation process, to
establish the extent in which a particular design and
implementation meets the set of specified security
• Certification is the endorsement that the
system/application meets their functional and security
requirements. It is the comprehensive technical analysis
of the security features and safeguards of a system to
establish the extent to which the security requirements
are satisfied.
Certification
• Certification uses a combinations of security
evaluation techniques:
- Risk analysis.
- Validation, verification and testing
- Security countermeasure evaluation
- Audit
• Certification should consider the following
issues:
- Security modes of operation
- Specific users and their training
- System and facility configuration and location
- Intercommunication with other systems
Vulnerabilities of Certification
• Organizations and users cannot count on the
certified product being free of security flaws.
Because new vulnerabilities are always being
discovered, no product is ever completely secure.
• Most software products must be securely
configured to meet certain protection
mechanisms.
• Certifications are not the definitive answer to
security. IS security depends on more than just
technical software protection mechanisms, such
as personnel and physical security measures.
Accreditation
• Accreditation is a formal declaration by a
Designated Approving Authority (DAA)
where an information system is approved to
operate in a particular security mode using a
prescribed set of safeguards at an acceptable
level of risk
• Accreditation is the official management
decision to operate a system. It is
management’s formal approval of the
adequacy of a system’s
Certification and Accreditation
• Accreditation looks at the following items:
- Particular security mode
- Prescribed set of countermeasures
- Defined threats; stated vulnerabilities
- Given operational concept and environment
- Stated interconnections to other systems
- Risk formally accepted
- Stated period of time
• Certification and accreditation should be an ongoing
process. A formal recertification and reaccreditation is
required whenever a major change occurs, a major
application is added, the security environment changes,
or significant technology is upgraded.
Examples of Guidelines for Certification
• Defense Information Technology Security
Certification and Accreditation Process
(DITSACP). DOD #5200.40 (December 1997)
• National Information Assurance Certification and
Accreditation Process (NIACAP). NSA #NSTISSI
No. 100. (April 2000).
• Guide for the Security Certification and
Accreditation of Federal Information Systems,
NIST SP 800-37. (October 2002)