Skip to main content

Über dieses Buch

As Grids and service-oriented architectures have evolved to a common infrastructure for providing and consuming services in research and commercial environments, mechanisms are needed to agree on the objectives and the quality of such service provision. There is a clear trend to use electronic contracts between service consumers and one or more service providers, in order to achieve the necessary reliability and commitment from all parties. Service Level Agreements (SLAs) are the means to model and manage such contracts in a unified way.

Grids and Service-Oriented Architectures for Service Level Agreements, the thirteenth volume of the CoreGRID series, contains current research and up-to date solutions from research and business communities presented at the IEEE Grid 2009 Workshop on Service Level Agreements in Grids, and the Service Level Agreements in Grids Dagstuhl Seminar 2009. The contributions in this volume cover Grid environments, but also generic models for SLA management that are applicable to service-oriented systems in general, like market-economic strategies, negotiation models, or monitoring infrastructures.

Grids and Service-Oriented Architectures for Service Level Agreements is designed for a professional audience composed of researchers and practitioners within the Grid community industry, and is also suitable for advanced-level students in computer science.



Monitoring Service Level Agreements in Grids with support of a Grid Benchmarking Service

As Computational Grids become more popular, a new generation of grid applications emerges, demanding strict and increasingly sophisticated guarantees of quality of service. This challenge has motivated the development of numerous technologies to enable service providers and consumers to establish Service Level Agreements (SLAs). The implementation of SLAs requires mechanisms for these agreements to be monitored and enforced, so that they can be dependable.
Most of existing SLA monitoring techniques are embedded in particular SLA specification, negotiation and management mechanisms. This poses significant limitations for their adoption in large scale, heterogeneous, decentralized grid infrastructures.
In this paper, we present how SLAs can be assessed, monitored and enforced with support of Jawari, a multi-platform, extensible and free of charge grid benchmarking service. Jawari works as an independent external entity that validates the adherence of the grid components to committed SLAs, by simply using the grid services just like an end-user would do. Doing so, it is able to observe the actual levels of quality of service the end-users are likely to experience.
Its benchmarks represent classes of grid applications with distinct requirements that expose the grid services to scenarios where the SLA is expected to be observed. Complementarily, SLA violation conditions can be routinely checked, and proper actions taken in response.
Ely de Oliveira, Franz-Josef Pfreundt

Reactive Monitoring of Service Level Agreements

Service Level Agreements require a monitoring system that checks that no party violates the agreement. Current monitoring techniques either have a high performance overhead or are not reliable enough. This paper proposes a new hybrid monitoring system that we call reactive monitoring. It tries to balance the disadvantages of established monitoring techniques, in particular online and offline monitoring. Online monitoring has a relatively high performance overhead and offline monitoring does not identify all possible violations.
Reactive monitoring combines online monitoring, which is used for reactively checking continuous SLA properties with a new passive monitoring scheme. This scheme is used for monitoring discrete SLA properties. It is based on cryptographic primitives that provide proof that either a certain stage in an interaction has been reached correctly with all participants in compliance of the service level agreements or that a violation has occurred. In the latter case the violating party can be identified.
A theoretical analysis shows that in the worst case scenario this new approach has the same overhead as online monitoring techniques and in most cases the overhead will be significantly lower.
Dalia Khader, Julian Padget, Martijn Warnier

Lessons Learned from Implementing WS-Agreement

WS-Agreement describes a protocol and structure for creating and representing service level agreements. In order to remain domain independent, the authors of the WS-Agreement specification have provided many extension points for domain specific content. This creates high degrees of freedoms for programmers to implement the specification. Many attempts to do this have been made in the past. In this paper, we explain what we have learned from our own and other projects’ attempts of implementing WS-Agreement. The paper presents a set of guidelines how the features of WS-Agreement can be used in a sound way that allows transferring large parts of the WS-Agreement logic into a generic and domain-independent WS-Agreement framework.
Dominic Battré, Matthias Hovestadt, Oliver Wäldrich

SLA-aware Resource Management

The management of infrastructure resources in a large-scale environment such as Grid Computing is a challenging task and places significant demands on resource discovery, scheduling and the underlying communication channels. The fulfillment of the business goals and service quality in such an environment requires an infrastructure to cope with changes in demand and infrastructure performance. In this paper, we propose an abstract service-oriented framework for SLA-aware dynamic resource management. The framework provides selfmanaging, self-configuration and self-healing strategies in order to support autonomic and ambient service management. We study an SLA negotiation process at the infrastructure resource layer, live migration for resource re-provisioning, a multi-layer architecture framework to monitor infrastructure resources and a harmonized interface to access arbitrary sources of infrastructure resources based on SLA requirements. Resource usage will be optimized according to the provider policies and SLA requirements.
Yih Leong Sun, Ron Perrott, Terence J Harmer, Christina Cunningham, Peter Wright, John Kennedy, Andy Edmonds, Victor Bayon, Jacek Maza, Gregor Berginc, Primož Hadalin

Distributed Trust Management for Validating SLA Choreographies

For business workflow automation in a service-enriched environment such as a grid or a cloud, services scattered across heterogeneous Virtual Organizations (VOs) can be aggregated in a producer-consumer manner, building hierarchical structures of added value. In order to preserve the supply chain, the Service Level Agreements (SLAs) corresponding to the underlying choreography of services should also be incrementally aggregated. This cross-VO hierarchical SLA aggregation requires validation, for which a distributed trust system becomes a prerequisite. Elaborating our previous work on rule-based SLA validation, we propose a hybrid distributed trust model. This new model is based on Public Key Infrastructure (PKI) and reputation-based trust systems. It helps preventing SLA violations by identifying violation-prone services at service selection stage and actively contributes in breach management at the time of penalty enforcement.
Irfan Ul Haq, Rehab Alnemr, Adrian Paschke, Erich Schikuta, Harold Boley, Christoph Meinel

Evaluation of Service Level Agreement Approaches for Portfolio Management in the Financial Industry

The idea of service-oriented Grid computing seems to have the potential for fundamental paradigm change and a new architectural alignment concerning the design of IT infrastructures. There is a wide range of technical approaches from scientific communities which describe basic infrastructures and middlewares for integrating Grid resources in order that by now Grid applications are technically realizable. Hence, Grid computing needs viable business models and enhanced infrastructures to move from academic application right up to commercial application. For a commercial usage of these evolutions service level agreements are needed. The developed approaches are primary of academic interest and mostly have not been put into practice. Based on a business use case of the financial industry, five service level agreement approaches have been evaluated in this paper. Based on the evaluation, a management architecture has been designed and implemented as a prototype.
Tobias Pontz, Manfred Grauer, Roland Kuebert, Axel Tenschert, Bastian Koller

Expressing Intervals in Automated Service Negotiation

During automated negotiation of services between autonomous agents, utility functions are used to evaluate the terms of negotiation. These terms often include intervals of values which are prone to misinterpretation. It is often unclear if an interval embodies a continuum of real numbers or a subset of natural numbers. Furthermore, it is often unclear if an agent is expected to choose only one value, multiple values, a sub-interval or even multiple sub-intervals. Additional semantics are needed to clarify these issues. Normally, these semantics are stored in a domain ontology. However, ontologies are typically domain specific and static in nature. For dynamic environments, in which autonomous agents negotiate resources whose attributes and relationships change rapidly, semantics should be made explicit in the service negotiation. This paper identifies issues that are prone to misinterpretation and proposes a notation for expressing intervals. This notation is illustrated using an example in WS-Agreement.
Kassidy P. Clark, Martijn Warnier, Sander van Splunter, Frances M.T. Brazier

GreenIT Service Level Agreements

In this paper we are introducing a framework towards the inclusion of Green IT metrics as part of service level agreements for future Grids and Clouds. As part of this effort we need to revisit Green IT metrics and proxies that we consider optimizing against in order to develop GreenIT as a Services (GaaS) that can be reused as part of a Software as a Service (SaaS) and Infrastructure Infrastructureas a service (IaaS) framework. We report on some of our ongoing efforts and demonstrate how we already achieve impact on the environment with our services.
Gregor von Laszewski, Lizhe Wang

Extending WS-Agreement with Multi-round Negotiation Capability

The WS-Agreement specification of the Open Grid Forum defines a language and a protocol for advertising the capabilities of service providers and creating agreements based on templates, and for monitoring agreement compliance at runtime. While the specification, which currently is in the process of transition from a proposed recommendation of the Open Grid Forum to a full recommendation, has been widely used after the initial publication in May 2007, it became obvious that the missing possibility to negotiate an agreement rather than just accepting an offer is limiting or inhibiting the use of WS-Agreement for a number of use-cases. Therefore, the Grid Resource Allocation Agreement Working Group of the Open Grid Forum started in 2008 to prepare an extension of WS-Agreement that adds negotiation capabilities without changing the current specification in a way, which leads to an incompatible new version of WS-Agreement. In this paper we present the results of this process with an updated version of the specification in mind and the first implementation in the European project SmartLM.
Angela Rumpl, Oliver Wäldrich, Wolfgang Ziegler

Enabling Open Cloud Markets Through WS-Agreement Extensions

Research into computing resource markets has mainly considered the question of which market mechanisms provide a fair resource allocation. However, while developing such markets, the definition of the unit of trade (i.e. the definition of resource) has not been given much attention. In this paper, we analyze the requirements for tradable resource goods. Based on the results, we suggest a detailed goods definition, which is easy to understand, can be used with many market mechanisms, and addresses the needs of a Cloud resource market. The goods definition captures the complete system resource, including hardware specifications, software specifications, the terms of use, and a pricing function. To demonstrate the usefulness of such a standardized goods definition, we demonstrate its application in the form of a WS-Agreement template for a number of market mechanisms for commodity system resources.
Marcel Risch, Jörn Altmann

Service Mediation and Negotiation Bootstrapping as First Achievements Towards Self-adaptable Cloud Services

Nowadays, novel computing paradigms as for example Cloud Computing are gaining more and more on importance. In case of Cloud Computing users pay for the usage of the computing power provided as a service. Beforehand they can negotiate specific functional and non-functional requirements relevant for the application execution. However, providing computing power as a service bears different research challenges. On one hand dynamic, versatile, and adaptable services are required, which can cope with system failures and environmental changes. On the other hand, human interaction with the system should be minimized. In this chapter we present the first results in establishing adaptable, versatile, and dynamic services considering negotiation bootstrapping and service mediation achieved in context of the Foundations of Self-Governing ICT Infrastructures (FoSII) project. We discuss novel meta-negotiation and SLA mapping solutions for Cloud services bridging the gap between current QoS models and Cloud middleware and representing important prerequisites for the establishment of autonomic Cloud services.
Ivona Brandic, Dejan Music, Schahram Dustdar

SLA Negotiation for VO Formation

Resource management systems are changing from localized resources and services towards virtual organizations (VOs) sharing millions of heterogeneous resources across multiple organizations and domains. The virtual organizations and usage models include a variety of owners and consumers with different usage, access policies, cost models, varying loads, requirements and availability. The stakeholders have private utility functions that must be satisfied and possibly maximized.
This paper proposes automated negotiation techniques between web services for the formation of virtual organizations. More specifically, a multi-issue sealed bid auction is implemented between a VO manager and potential VO members on the resources they provide or on their payment for requested resources. We evaluate our approach to show that negotiation allows to form a more efficient VO.
Shamimabi Paurobally

From Service Markets to Service Economies – An infrastructure for protocol-generic SLA negotiations

Visions of 21st century’s information systems show highly specialized digital services and resources, interacting continuously and with a global reach. For a broad adoption of this vision in a commercial context it is crucial to have a mechanism in place to guarantee quality of service and to decentrally coordinate the involved resources. Current service infrastructures try to tackle these problems by applying socioeconomic mechanisms such as electronic negotiations and service level agreements. Such technologies allow for the implementation of electronic service markets in analogy to real-world markets for everyday goods. However, economic theory claims that different market situations and negotiated products (i.e. SLAs) demand different negotiation protocols in order to reach the highest-possible overall efficiency of the system. Thus we argue that next generation service infrastructures will be based on a global service economy where several different service markets and thus protocols are present at any given point in time. In this paper we present a novel approach for such an infrastructure, based on structured protocol descriptions and software-agent technology.
Sebastian Hudert

Service Level Agreements in BREIN

With electronic business (eBusiness) becoming ubiquitous, the traditional ways of doing commerce need to be changed or completely replaced to support the end users effectively in performing their business. This includes especially the representation of business relationships with an electronic format to allow for automated processing of the respective parts of e.g. contractual obligations. One prominent representation tool are Service Level Agreements. Conceptually established as paper representation to describe parts of contracts of telecom operators, SLAs have become a research topic in the ICT domain now since several years.
However, the current State of the Art in Service Level Agreements and their management, still shows several deficits, which prevented the uptake of eBusiness solutions (based on Service Level Agreements) so far. This paper will present how the BREIN project enhanced, amongst others Service Level Agreement Management with capabilities from the Multiagent and Semantic domain, to provide an enhanced solution, compared to existing technologies. Thereby the main emphasize was on basing the developments on existing results to concentrate on gap filling instead of re-invention of the wheel.
Bastian Koller, Henar Munoz Frutos, Giuseppe Laria

Negotiation and Monitoring of Service Level Agreements

Service level agreements (SLAs) provide a means to define specific Quality of Service (QoS) guarantees between providers and consumers of services. Negotiation and definition of these QoS characteristics is an area of significant research. However, defining the actions that take place when an agreement is violated is a topic of more recent focus. This paper discusses recent advances in this field and propose some additional features that can help both consumers and producers during the enactment of services. These features include the ability to (re)negotiate penalties in an agreement, and specifically focuses on the renegotiation of penalties during enactment to reflect ongoing violations.
Thomas B. Quillinan, Kassidy P. Clark, Martijn Warnier, Frances M.T. Brazier, Omer Rana


Weitere Informationen

Premium Partner