Cost-Based Optimization of Service Compositions

Abstract

For providers of composite services preventing cases of SLA violations is crucial. Previous work has established runtime adaptation of compositions as a promising tool to achieve SLA conformance. However, in order to get a realistic and complete view of the decision process of service providers, the costs of adaptation need to be taken into account. In this paper we formalize the problem of finding the optimal set of adaptations, which minimizes the total costs arising from SLA violations and the adaptations to prevent them. We present possible algorithms to solve this complex optimization problem, and detail an end-to-end system based on our earlier work on the PREvent (prediction and prevention based on event monitoring) framework, which clearly indicates the usefulness of our model. We discuss experimental results that show how the application of our approach leads to reduced costs for the service provider, and explain the circumstances in which different algorithms lead to more or less satisfactory results. Abstract-For providers of composite services preventing cases of SLA violations is crucial. Previous work has established runtime adaptation of compositions as a promising tool to achieve SLA conformance. However, in order to get a realistic and complete view of the decision process of service providers, the costs of adaptation need to be taken into account. In this paper we formalize the problem of finding the optimal set of adaptations, which minimizes the total costs arising from SLA violations and the adaptations to prevent them. We present possible algorithms to solve this complex optimization problem, and detail an end-to-end system based on our earlier work on the PREvent (prediction and prevention based on event monitoring) framework, which clearly indicates the usefulness of our model. We discuss experimental results that show how the application of our approach leads to reduced costs for the service provider, and explain the circumstances in which different algorithms lead to more or less satisfactory results.

The cost for meeting SLA dependability requirements; implications for customers and providers

“…− C c|n,m,ω P n,m,ω|λ,α,β (τ) (19) Note that the deployment cost in (19) is related to the price for the services delivered by underlying systems reduced with the expected compensations.…”

“…In [18] a frame work is proposed for modelling the optimal investment for availability for a two state semi-Markov modelled system where the customer compensation is depending on the down time duration and its variance. A simi lar trade-off is formalized as an optimization problem in [19] with the objective to minimize the total cost of system improvements and compensations. A two-state Markov model is analysed in [20] where an upper bound for an insurance premium is calculated based upon the dependability related SLOs number of failures, cumu lative down time duration and number of down times longer than a defined threshold.…”

Towards Compensable SLAs

“…Regarding the properties fulfilled by the compensations of the real word SLAs, we highlight that GNWT Sample 4 of Figure 3 depicts an inconsistent compensation function; this may represent a mistake in the SLA derived from the usage of natural language. With regards to the examples found in the related work, it is remarkable that the proposal of Leitner et al in [5] formalizes the problem of finding the optimal set of adaptations, which minimizes the total costs arising from SLA violations and the adaptations to prevent them. In this work, a model for penalty functions is presented; this formalization has been the starting point of our motivational scenario description presented in Section 2 and consequently, our approach represents an extension to this model in order to develop a complete formalization for Compensable Guarantees and SLAs.…”

Introducing the Vienna Platform for Elastic Processes

“…Through the executor, the WfM is also able to measure the service response time. This functionality is needed in order to identify deviations from the expected service behavior at an early point of time and avoid breaches of SLAs, which can lead to penalties [15]. Response time includes the service invocation time as well as the network latency.…”

“…While such optimization approaches have also been used in QoS-aware service compositions, as presented, e.g., in [15,23,24,26], these composition algorithms do not take into account resources. Instead, they are based on QoS assurances given by the service providers.…”