Abstract—The availability of scarce resources in a serviceoriented system demands for context-aware selection policies that adapt based on service-level agreements (SLAs). One of the open issues is to prioritize service requests in dynamically changing environments where concurrent instances of processes may compete for resources. Here we propose a runtime monitoring approach to observe the actual state of the system. We argue that priorities should be assigned to requests based on potential violations of SLA objectives. While most existing work in the area of quality of service monitoring and SLA modeling focuses typically on purely technical systems, we consider service-oriented systems spanning both softwarebased services and human actors. Adaptive request scheduling in such systems is challenging due to the poorly predictable behavior of human actors in performing tasks. Our approach helps to cope with these challenges by prioritizing service requests that may cause violations of SLAs and corresponding objectives that are associated with processes. Keywords-service-oriented systems, monitoring, adaptation, service-level agreements, scheduling, human factors I.

Abstract—The availability of scarce resources in a serviceoriented system demands for context-aware selection policies that adapt based on service-level agreements (SLAs). One of the open issues is to prioritize service requests in dynamically changing environments where concurrent instances of processes may compete for resources. Here we propose a runtime monitoring approach to observe the actual state of the system. We argue that priorities should be assigned to requests based on potential violations of SLA objectives. While most existing work in the area of quality of service monitoring and SLA modeling focuses typically on purely technical systems, we consider service-oriented systems spanning both softwarebased services and human actors. Adaptive request scheduling in such systems is challenging due to the poorly predictable behavior of human actors in performing tasks. Our approach helps to cope with these challenges by prioritizing service requests that may cause violations of SLAs and corresponding objectives that are associated with processes. Keywords-service-oriented systems, monitoring, adaptation, service-level agreements, scheduling, human factors I.

Abstract. Existing research work considers runtime adaptation of service compositions as a viable tool to prevent violations of service level agreements. In previous work we have formalized the optimization problem of identifying the most suitable adaptations to prevent a predicted set of violations, and presented suitable algorithms to solve this problem. Here, we introduce the idea of stepwise optimization as a solution to the problem of how to deal with situations when the optimization result is not available in time, i.e., when decisions need to be taken before the optimization problem can be fully solved. 1

Providers of applications deployed in an Infrastructureas-a-Service cloud permanently face the decision of whether it is more cost-efficient to scale up (i.e., rent more resources from the cloud) or to delay incoming requests, even though doing so may lead to dissatisfied customers and broken service level agreements. This decision is further complicated by the fact that not all customers have the same agreements, and not all requests require the same amount of resources devoted to them. In this paper, we present an approach for optimally scheduling incoming requests to virtual computing resources in the cloud, so that the sum of payments for resources and loss incurred by service level agreement violations is minimized. We discuss our approach based on an illustrative use case. Furthermore, we present a numerical evaluation based on real-life request data, which shows that our agreementaware algorithm improves upon earlier work, which does not take service level agreements into account. I.