Logo des Repositoriums

Auflistung nach Schlagwort "Performance Prediction"

1 - 2 von 2
  • Zeitschriftenartikel
    Improving IaaS Cloud Analyses by Black-Box Resource Demand Modeling
    (Softwaretechnik-Trends Band 35, Heft 3, 2015) Groenda, Henning; Stier, Christian
    In Infrastructure as a Service (IaaS) Cloud scenarios, data center operators require specifications of Virtual Machine (VM) behavior for data center middle- and long-term planning and optimization. The planning is usually supported by simulations. While users can leverage white-box application knowledge, data center operators have to rely on metrics at the level of resource demands provided by virtualization and cloud middleware platforms. Existing simulations for data center planning do not combine both viewpoints and either require white-box knowledge or focus on short-term predictions using statistical estimators. Our approach allows modeling varying resource demand of black-box VMs based on the Descartes Load Intensity Model (DLIM). The black-box VM models are integrated in the SimuLizar performance simulator complementing the existing grey- and white-box models in order to improve reasoning on (de-) consolidation decisions.
  • Konferenzbeitrag
    Modeling and Simulation of Dynamic Containerized Software Architectures using Palladio
    (SE 2024 - Companion, 2024) Hagel, Nathan
    Nowadays, distributed applications are often not statically deployed on virtual machines. Instead, a desired state is defined declaratively. A control loop then tries to create the desired state in a cluster. To predict the impact on the performance of a system using these deployment techniques is difficult. This paper introduces a method to predict the performance impact of the usage of containers and container orchestration in the deployment of a system. Our proposed approach enables system simulation and experimentation with various mechanisms of container orchestration, including autoscaling and container scheduling using the Palladio-Component-Model (PCM). We validated this approach using a Kubernetes reference cluster which we modelled using a workflow defined in the authors bachelor thesis [Ha22] [Ha23]. Our findings suggest, that the most common concepts in container orchestration can be modelled and simulated using Palladio and the PCM-Extension of [Ha22].