Model-Based Performance Anticipation in Multi-tier Autonomic Systems: Methodology and Experiments

Nabila Salmi (1, 2) , Bruno Dillenseger (3) , Ahmed Harbaoui (4, 5) , Jean-Marc Vincent (5)

1 LISTIC - Laboratoire d'Informatique, Systèmes, Traitement de l'Information et de la Connaissance
2 LSI - Laboratoire des Systèmes Informatiques
3 Orange Labs [Meylan]
4 LIG - Laboratoire d'Informatique de Grenoble
5 MESCAL - Middleware efficiently scalable

This paper advocates for the introduction of perfor- mance awareness in autonomic systems. Our goal is to introduce performance prediction of a possible target configuration when a self-* feature is planning a system reconfiguration. We propose a global and partially automated process based on queues and queuing networks modelling. This process includes decomposing a distributed application into black boxes, identifying the queue model for each black box and assembling these models into a queuing network according to the candidate target configuration. Finally, performance prediction is performed either through simulation or analysis. This paper sketches the global process and focuses on the black box model identification step. This step is automated thanks to a load testing platform enhanced with a workload control loop. Model identification is based on statistical tests. The identified models are then used in performance prediction of autonomic system configurations. This paper describes the whole process through a practical experiment with a multi-tier application.

Mots clés: autonomic systems; performance evaluation

Domaines: Performance et fiabilité [cs.PF]; Génie logiciel [cs.SE]

Nabila Salmi, Bruno Dillenseger, Ahmed Harbaoui, Jean-Marc Vincent. Model-Based Performance Anticipation in Multi-tier Autonomic Systems: Methodology and Experiments. International Journal On Advances in Networks and Services, 2010, 3 (3-4), pp.346-360. ⟨hal-00993647⟩