Banca de DEFESA: BABACAR MANE

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
DISCENTE : BABACAR MANE
DATA : 21/12/2020
HORA: 14:00
LOCAL: web conferencia
TÍTULO:

 Evolving the interoperability from SaaS and DaaS/DBaaS: the MIDAS case


PALAVRAS-CHAVES:

Cloud Computing, Middleware, Interoperability, Metamodel


PÁGINAS: 132
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Ciência da Computação
SUBÁREA: Metodologia e Técnicas da Computação
ESPECIALIDADE: Sistemas de Informação
RESUMO:

The provider of both Data as a Service (DaaS) and Database as a Service (DBaaS) stores and manages a high volume of heterogeneous data. Such data are produced by mobile computing, ubiquitous devices, social networks, and they are distributed geographically

and available to consumers and organizations as services through an API. Users face a
challenge when accessing similar distributed data from distinct DaaS/DBaaS providers due to the lack of a standard API and tools. Consequently, cloud users face interoperability and integration issues for consumption, provisioning, management, and supervision resources among distinct clouds. In such a heterogeneous environment, organizations desiring to exchange their data among clouds or move their applications to distinct clouds will face a lock-in situation due to the lack of a standard solution. Middleware has been employed to deal with interoperability issues to minimize the effort to overpass lock-in problems. Thus, this thesis introduces the middleware MIDAS to minimize the effort to interoperate SaaS and DaaS. We can summarize three main contributions from this work. Initially, the MIDAS Middleware provides (i) syntactic interoperability among SaaS and DaaS. MIDAS allows a cloud query from SaaS to request data from distinct DaaS/DBaaS. Data attributes and conditions are described in SQL or MongoDB queries. Therefore, a SaaS request may be affected by the evolution of the DaaS attributes. As a second contribution, MIDAS was evolved to provide (ii) semantic interoperability to ensure DaaS consistency and maintain SaaS’s original query. Currently, MIDAS runtime implementations rely on Cloud Foundry, Amazon Web Services, OpenShift, and Heroku providers. Some cloud development and runtime environment diversities (.e.g., application framework and programming languages) prevent implementing, running, or deploying applications on a large scale. To avoid ambiguity in the development and seamless deployment of MIDAS in different cloud providers, we present the third contribution of this thesis, a Domain-Specific Modeling Language (DSML), which is (iii)  the metamodel of MIDAS architecture and a Unified Modeling Language (UML) profile. This metamodel should guide the instantiation of MIDAS platform-independent models and other implementations. It is validated by executing three types of user queries to measure the constructor’s coverage levels to define the middleware and implement it in a specific cloud provider.



MEMBROS DA BANCA:
Presidente - 1232218 - DANIELA BARREIRO CLARO
Interno - 2352062 - MARLO VIEIRA DOS SANTOS E SOUZA
Interno - 2049039 - VANINHA VIEIRA DOS SANTOS
Externo à Instituição - GLAUCO DE FIGUEIREDO CARNEIRO - UNIFACS
Externo à Instituição - JOSÉ MARIA NAZAR DAVID - UFJF
Notícia cadastrada em: 05/01/2021 14:17
SIGAA | STI/SUPAC - - | Copyright © 2006-2024 - UFBA