Systematically Translating Service Level Objectives Into Design and Operational Policies for Multi-Tier Applications

This paper proposes a systematic and practical approach that combines fine-grained performance modeling with regression analysis to translate Service Level Objectives (SLOs) into design and operational policies for multi-tier applications. These policies can then be used for designing a service to meet the SLOs and monitoring the service thereafter for violations at runtime. They demonstrate that the approach is practical and can be applied to commonly used multi-tier applications with different topologies and performance characteristics. Their approach handles both request-based and session-based workloads and deals with workload changes in terms of both request volume and transaction mix.