Abstract
In this chapter we present a survey of research work related to the semantic modelling of security, semantic SLA modelling, and the current state of the art in SLA-based system governance. Based on this survey, and after observing the essential aspects needed to semantically model an SLA, we first propose a semantic model of resource dependability. This model can be used to semantically encode in SLA the service commitments (to customers) and resource capacity (from suppliers) in terms of usage, performance, and other QoS characteristics that represent non-functional properties. On the basis of this model, we propose a flexible approach to SLA-based system governance that allows for elastic provisioning of resources (by autonomic processes) that meet NFP requirements. This approach can be used to monitor and manage services such that they meet (and continue to meet) agreed levels of QoS.
TopChapter Structure
This Chapter contains the following four parts:
- 1.
an analysis of relevant background;
- 2.
a proposed solution to semantic modelling of resource dependability;
- 3.
a flexible approach to SLA-based system governance; and
- 4.
a discussion of future directions.
In this Chapter, we present a survey of research work related to semantic modelling of security, semantic SLA modelling, and the current state of the art in SLA-based system governance. This Chapter provides an overview of existing approaches in two areas: (1) tools and approaches to model and analyse security, risk and vulnerabilities of IT landscapes and (2) SLA modelling, SLA negotiation and deployment. The former gives an understanding of the risk to which a provider of ICT services may be exposed, while the latter provides a means to model these and capture how the dynamics of the runtime system can indicate the presence of a threat and effectiveness of mitigation strategies.
The review is followed by a proposal for a semantic model of resource dependability. This approach aims to encode service commitments to customers and resource capacity in a semantic driven way.
On the basis of this model, we then propose a framework for elastic service provisioning that allows services to be deployed in such a way that they can be managed to meet the stated QoS requirements.
The Chapter concludes with a list of future research directions.
TopBackground
In a typical security modelling approach, a system is described as a collection of assets or resources, which typically include computers, networks and data. These have properties such as confidentiality, integrity and availability, which are important to the organisation operating the system. They also have vulnerabilities whereby these properties can be undermined.
We need a more general set of dependability attributes to express the behaviour of resources even if accessed across organisational boundaries, and to manage their run-time composition to meet (potentially changing) requirements. Moreover, the collection of resources will itself change as inter-organisational relationships are formed and broken, or discarded as untrustworthy.
Key Terms in this Chapter
Service: A means for a service provider to receive, process and respond to requests from consumers.
Quality of Service (QoS): The level of service provided by a service in processing and responding to requests.
Service Governance: The processes and mechanisms needed to coordinate and manage the use of IT-related activities and resources to meet the operative and strategic goals of the company providing the services.
Resource: An asset (e.g. another service) used by one (or more) services when processing and responding to requests.
Quality of Experience (QoE): The level of service experienced by a consumer in its use of one (or more) services.
Service Level Agreement (SLA): A bilateral agreement that encodes the terms of the business relationship between a service provider and a customer.
Consumer: Someone that sends requests to one (or more) services.
Customer: Someone who is responsible for requests sent to one (or more) services.