Automated Workload Generation for Testing Elastic Web Applications - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2016

Automated Workload Generation for Testing Elastic Web Applications

Résumé

Web applications are often exposed to unpredictable workloads, which make infrastructure resource management difficult. Resource may be overused when the workload is high and underused when the workload is low. A solution to deal with unpredictable workloads is to migrate web applications to cloud computing infrastructures, where resources vary according to demand. Since resource variations happen during the application life cycle, adaptation tasks must be performed at runtime. The resource variation and the adaptation tasks lead web applications to different states that do not exist in non-elastic infrastructure, which we call elasticity states. We claim that elasticity states may reveal supplementary application errors and that web applications must be tested accordingly. For that, web applications must be lead through elasticity states throughout the test. The natural way to lead web applications through elasticity states is to expose them to workload variations. However, generating the correct workload to induce infrastructure adaptation in a minimal time, without overloading the application, is a difficult task. Aiming to make the workload generation more efficient, we propose a two phases approach that first analyzes the application adaptation behavior and then generates the appropriate workload. We validated our approach by conducting several experiments on Google and Amazon cloud infrastructures. In these experiments, a web application was successfully conducted through the predefined resource variations.
Fichier principal
Vignette du fichier
sbac-albonico-mottu-sunye-version2.pdf (341.08 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01317723 , version 1 (18-05-2016)

Identifiants

  • HAL Id : hal-01317723 , version 1

Citer

Michel Albonico, Jean-Marie Mottu, Gerson Sunyé. Automated Workload Generation for Testing Elastic Web Applications. 2016. ⟨hal-01317723⟩
248 Consultations
222 Téléchargements

Partager

Gmail Facebook X LinkedIn More