Doctoral Thesis Defense - Clayton Maciel Costa

Date: 
Friday, September 29, 2017, 2:30pm

Title: Efficient adaptive query processing on large database systems available in the cloud environment

Abstract: Nowadays, many companies are migrating their applications and data to cloud service providers, mainly because of their ability to answer quickly to business requirements. Thereby, the performance is an important requirement for most customers when they wish to migrate their applications to the cloud. Therefore, in cloud environments, resources should be acquired and releasedautomatically and quickly at runtime. Moreover, the users and service providers expect to get answers in time to ensure the service SLA (Service Level Agreement). Consequently, ensuring the QoS (Quality of Service) is a great challenge and it increases when we have large amounts of data to be manipulated in this environment.
To resolve this kind of problems, several researches have been focused on shorter execution time using adaptive query processing and/or prediction of resources based on current system status. However, they present important limitations. For example, most of these works does not use monitoring during query execution and/or presents intrusive solutions, i.e. applied to the particular context.
The aim of this thesis is the development of new solutions/strategies to efficient adaptive query processing on large databases available in a cloud environment. It must integrate adaptive re-optimization at query runtime and their costs are based on the SRT (Service Response Time – SLA QoS performance parameter). Finally, the proposed solution will be evaluated on large scale with large volume of data, machines and queries in a cloud computing infrastructure.
Finally, this work also proposes a new model to estimate the SRT for different request types (database access requests). This model will allow the cloud service provider and its customers to establish an appropriate SLA relative to the expected performance of the services available in the cloud.

Jury: - Rector of the University of Minho

         - Doctor Carlos Nuno Bizarro e Silva Laranjeiro, Assistant Professor - Departamento de Engenharia Informática da Faculdade de Ciências e Tecnologia da Universidade de Coimbra

         - Doctor Miguel Ângelo Marques de Matos, Assistant Professor - Departamento de Engenharian Informática do Instituto Superior Técnico da Universidade de Lisboa

         - Doctor José Orlando Roque Nascimento Pereira, Assistant Professor - Departamento de Informática da Escola de Engenharia da Universidade do Minho

         -Doctor António Luís Pinto Ferreira Sousa, Assistant Professor - Departamento de Informática da Escola de Engenharia do Universidade do Minho

Location: Auditorium B1 - CPII - Campus de Gualtar - Universidade do Minho - Braga

Photos: