User-centric Adaptation Analysis of Multi-tenant Services
Multi-tenancy is a key pillar of cloud services. It allows different users to share computing and virtual resources transparently, meanwhile guaranteeing substantial cost savings. Due to the tradeoff between scalability and customization, one of the major drawbacks of multi-tenancy is limited config...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2016 |
| País: | España |
| Institución: | Universidad de Sevilla (US) |
| Repositorio: | idUS. Depósito de Investigación de la Universidad de Sevilla |
| OAI Identifier: | oai:idus.us.es:11441/128280 |
| Acceso en línea: | https://hdl.handle.net/11441/128280 https://doi.org/10.1145/2790303 |
| Access Level: | acceso abierto |
| Palabra clave: | User systems Human information processing Multi-tenant services |
| Sumario: | Multi-tenancy is a key pillar of cloud services. It allows different users to share computing and virtual resources transparently, meanwhile guaranteeing substantial cost savings. Due to the tradeoff between scalability and customization, one of the major drawbacks of multi-tenancy is limited configurability. Since users may often have conflicting configuration preferences, offering the best user experience is an open challenge for service providers. In addition, the users, their preferences, and the operational environment may change during the service operation, thus jeopardizing the satisfaction of user preferences. In this article, we present an approach to support user-centric adaptation of multi-tenant services. We describe how to engineer the activities of the Monitoring, Analysis, Planning, Execution (MAPE) loop to support user-centric adaptation, and we focus on adaptation analysis. Our analysis computes a service configuration that optimizes user satisfaction, complies with infrastructural constraints, and minimizes reconfiguration obtrusiveness when user- or service-related changes take place. To support our analysis, we model multitenant services and user preferences by using feature and preference models, respectively. We illustrate our approach by utilizing different cases of virtual desktops. Our results demonstrate the effectiveness of the analysis in improving user preferences satisfaction in negligible time. |
|---|