際際滷

際際滷Share a Scribd company logo
Trust Assessment for Cloud Identity Providers
Using Analytical Hierarchy Process
Brian Cusack
Eghbal Ghazi Zadeh
Objective
 Develop a trust management framework for cloud
identity customers that:
 Aggregates, manages and measures trust-related
information
 Attributes, Characteristics, Metrics, Secure elements
 From different sources which are available and relevant
while assessing the trustworthiness of a cloud identity
providers
2
Cloud Computing
3
 XaaS (Anything as a Service)
 TaaS (Trust as a Service)
 IDaaS (Identity as a Service)
Cloud Computing
 Every cloud service has a method of managing
identities that may address some of challenges but a
user requires to know the trust that may be put in
the service.
 The techniques and methods used can be measured
by the trust measurement system to help a user to
make a good decision
4
Cloud Identity Management System
 OAuth
 SAML
 Windows CardSpace, Info Card
 Higgins
 U-Prove
 Identity Mixer (Idemix)
 OpenID
 OpenID Connect
5
OpenID Connect Workflow
6
Auth Server
Auth Server
Auth Server
Auth Server
Auth Server
Auth Server
Cloud Service Provider
Cloud Identity Provider
Cloud Identity User
Trust Framework
Problem Statement
 Between CIdPs and CIdUs, what are the
Essential System Characteristics (ESC) of the
published trust establishment methods?
 Prioritizing characteristics and attributes
 Multi-Criteria Decision Making (MCDM)
 Analytical Hierarchy Process (AHP)
7
Service Trust Evaluation System Architecture
8
AHP
9
 AHP approach is one of the more extensively
used MCDM methods.
 decomposition, or the hierarchy construction.
 comparative judgments, or defining and
executing data collection to obtain pairwise
comparison data on elements of the
hierarchical structure.
 synthesis of priorities, or constructing an
overall priority rating.
ESC OF CLOUD IDENTITY PROVIDERS
 Balancing
 Single sign-on
 Lifecycle
 Privacy
 Risk
 Standards
10
Hierarchy of CIdP Decision Making
11
Weights for Main Characteristics
12
AHP Result for Standard
13
Conclusion
 Selecting a trustable and Secure CIdP is
essential
 Users require trust information
 Selection of a CIdPs as an MCDM problem
 Critical criteria affecting the decision making
 Theoretical three level AHP model

14
Thank you
15

More Related Content

Trust assessment 2017 for cloud identity providers using analytical hierarchical process

  • 1. Trust Assessment for Cloud Identity Providers Using Analytical Hierarchy Process Brian Cusack Eghbal Ghazi Zadeh
  • 2. Objective Develop a trust management framework for cloud identity customers that: Aggregates, manages and measures trust-related information Attributes, Characteristics, Metrics, Secure elements From different sources which are available and relevant while assessing the trustworthiness of a cloud identity providers 2
  • 3. Cloud Computing 3 XaaS (Anything as a Service) TaaS (Trust as a Service) IDaaS (Identity as a Service)
  • 4. Cloud Computing Every cloud service has a method of managing identities that may address some of challenges but a user requires to know the trust that may be put in the service. The techniques and methods used can be measured by the trust measurement system to help a user to make a good decision 4
  • 5. Cloud Identity Management System OAuth SAML Windows CardSpace, Info Card Higgins U-Prove Identity Mixer (Idemix) OpenID OpenID Connect 5
  • 6. OpenID Connect Workflow 6 Auth Server Auth Server Auth Server Auth Server Auth Server Auth Server Cloud Service Provider Cloud Identity Provider Cloud Identity User Trust Framework
  • 7. Problem Statement Between CIdPs and CIdUs, what are the Essential System Characteristics (ESC) of the published trust establishment methods? Prioritizing characteristics and attributes Multi-Criteria Decision Making (MCDM) Analytical Hierarchy Process (AHP) 7
  • 8. Service Trust Evaluation System Architecture 8
  • 9. AHP 9 AHP approach is one of the more extensively used MCDM methods. decomposition, or the hierarchy construction. comparative judgments, or defining and executing data collection to obtain pairwise comparison data on elements of the hierarchical structure. synthesis of priorities, or constructing an overall priority rating.
  • 10. ESC OF CLOUD IDENTITY PROVIDERS Balancing Single sign-on Lifecycle Privacy Risk Standards 10
  • 11. Hierarchy of CIdP Decision Making 11
  • 12. Weights for Main Characteristics 12
  • 13. AHP Result for Standard 13
  • 14. Conclusion Selecting a trustable and Secure CIdP is essential Users require trust information Selection of a CIdPs as an MCDM problem Critical criteria affecting the decision making Theoretical three level AHP model 14

Editor's Notes

  1. Cloud entities: This component is responsible for interaction with customers and understanding their application needs. It performs discovery and ranking of services using other components such as trust management, direct/indirect trust, and evaluation methods. Monitoring and history information: This component discovers services that can satisfy user requirements, then it closely monitors the trust performance of the service. The related history records are stored in the service database for access and reference. Computing service network structure and catalogue: This component builds the network structure and the features advertised by the various different providers. The computing resources are divided into different classes to satisfy demand distribution.
  2. Cloud entities: This component is responsible for interaction with customers and understanding their application needs. It performs discovery and ranking of services using other components such as trust management, direct/indirect trust, and evaluation methods. Monitoring and history information: This component discovers services that can satisfy user requirements, then it closely monitors the trust performance of the service. The related history records are stored in the service database for access and reference. Computing service network structure and catalogue: This component builds the network structure and the features advertised by the various different providers. The computing resources are divided into different classes to satisfy demand distribution.