Publication: Modelo de interoperabilidad para plataformas de cloud computing basado en tecnologías del conocimiento
Loading...
Identifiers
Publication date
2013-10
Defense date
2013-11-25
Authors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
La Internet de los servicios ha ido modificando la Web desde un mero repositorio de
información hasta convertirse en una plataforma para servicios y transacciones donde las
organizaciones ofertan servicios de todo tipo, incrementando sus procesos de negocio
exponiéndolos a través de la Web. Unido a esta transformación surgen nuevos paradigmas,
como el Software-as-a-Service y el Cloud Computing, cuyos modelos se adaptan a esta nueva
concepción y prometen crear nuevos niveles de eficiencia mediante la compartición de
funcionalidades y recursos a gran escala. Con todo ello, las tecnologías relacionadas con el
Cloud Computing han ido incrementando su presencia e importancia en el mundo de las
tecnologías de la información y han evolucionado hasta convertirse en un conjunto maduro de
innovaciones tecnológicas capaces de proporcionar una sólida infraestructura para el
paradigma SaaS. El Cloud Computing y el SaaS abren las puertas a economías de gran escala y
también a nuevos horizontes pero se encuentran con un importante número de retos que
afrontar entre los que se encuentran:
La falta de modelos contrastados para determinar bajo qué condiciones es rentable
para las organizaciones el proceso de migración hacia estos modelos.
La falta de métodos probados (por ejemplo, guías de arquitecturas) para facilitar dicha
migración.
La falta de métodos de integración a diferentes niveles, hecho que dificulta el proceso
de desarrollo y entrega de un software capaz de comunicarse con otros elementos de
la plataforma cloud, perdiendo una de sus características más importantes.
Relacionado con el último aspecto comentado se encuentra el concepto de la
interoperabilidad. La interoperabilidad siempre se ha considerado como una de las asignaturas
pendientes de los sistemas de información, ya que la heterogeneidad de los sistemas eleva la
complejidad de los medios necesarios para alcanzarla. Este problema se está acrecentando con
la creciente proliferación de sistemas de información construidos de forma independiente que
dificultan o impiden los intercambios de información entre ellos. Debido a esto se pierde una
de las propiedades más interesantes de los sistemas de computación en nube pero además se
limita el avance hacia la posibilidad de un mundo completamente interoperable donde el
acceso a la información sería prácticamente infinito.
El problema de la interoperabilidad se ha afrontado desde diversas perspectivas, pero resultan
especialmente interesantes los enfoques realizados por las tecnologías semánticas, donde
alcanzar la interoperabilidad a distintos niveles es históricamente uno de sus objetivos
principales. La Web Semántica fue concebida principalmente para solucionar los problemas de
sobrecarga y heterogeneidad de los datos en la Web, hecho que trae como consecuencia la
ausencia de interoperabilidad. Su enfoque es el de añadir semántica a los datos de manera que
las máquinas sean capaces de procesarlos, razonar con ellos, combinarlos y realizar
deducciones lógicas de manera muy similar a como lo haría un ser humano, proporcionando
para ello una serie de herramientas útiles.
Dentro de este ámbito, esta tesis doctoral reúne los conceptos más relevantes y emergentes
en la actualidad como Cloud Computing, SaaS, tecnologías semánticas, modelado de procesos
de negocio, interoperabilidad entre sistemas, etc. El objetivo es fomentar la evolución de
nuevas plataformas orientadas en torno a la interoperabilidad y la reducción de costes que
puede suponer un impacto significativo en la industria. La investigación realizada, toma estas
tecnologías y paradigmas como base para poder definir un estándar o lenguaje común que
permita la interoperabilidad entre aplicaciones dentro de un entorno de Cloud Computing. Con
esto se consigue dar un paso importante en las áreas de las tecnologías mencionadas,
suponiendo un enfoque claramente innovador y cuyos resultados resultarán de gran
relevancia dentro del ámbito de los sistemas de información.
Por todo ello, la solución enmarcada dentro de esta tesis doctoral está orientada al diseño de
un lenguaje semántico basado en ontologías y de un sistema global que trabaja a su alrededor
capaz de capturar el conocimiento de las aplicaciones alojadas en una estructura de Cloud
Computing habilitando la realización de procesos de intercambio de información entre
aplicaciones heterogéneas de forma automática. Esta tesis doctoral plantea, por tanto, la
combinación del Cloud Computing y la Semántica para afrontar el problema de la
interoperabilidad a nivel de aplicación. Este enfoque no se encuentra extendido en la
actualidad debido a la “reciente” explosión de los sistemas cloud, pero los beneficios que
pueden aportarse mutuamente suponen una gran motivación para trabajar e investigar en
profundidad sobre ello.
La metodología de investigación seguida en esta tesis doctoral para alcanzar los objetivos y
demostrar las hipótesis planteadas ha consistido en:
Estudio del estado de arte. Mediante el estudio de los trabajos relacionados se pueden
conocer los métodos y tecnologías necesarios para alcanzar los objetivos planteados.
Este estudio se ha dividido en cinco bloques principales:
o Cloud Computing.
o Web Semántica.
o Servicios Web.
o Servicios Web Semánticos.
o Interoperabilidad.
Análisis de los aspectos más representativos de la literatura. El análisis de los
diferentes conceptos tratados en el estado del arte, permite realizar una toma de
decisiones correcta y sólidamente fundamentada para utilizar las tecnologías y
herramientas más adecuadas al problema.
Presentación de una primera aproximación de la solución. Esta etapa permitirá sentar
las bases y requisitos que se deben cumplir para cubrir los objetivos propuestos. Esta
parte de la investigación emplea las conclusiones extraídas del análisis previo y
permitirá concluir la viabilidad de la investigación y también si las herramientas
seleccionadas son suficientes para alcanzar la solución deseada.
Diseño final de la solución. En esta fase se concluye el proceso de diseño. Todos los
requisitos han de quedar cubiertos y debe presentar un nivel de detalle suficiente para
poder comenzar la implementación del entorno para su validación.
Implementación del lenguaje y configuración de la plataforma de Cloud Computing.
Esta fase marca el inicio de la validación que permitirá comprobar que el diseño
cumple con los objetivos marcados y que se encuentra en la línea de las hipótesis
propuestas. La configuración de la plataforma de computación en nube permitirá
evaluar y validar la investigación en un entorno real.
Validación. En esta fase se comprueba que toda la funcionalidad diseñada y
desarrollada permite representar el conocimiento relativo al problema y la
interoperabilidad de las aplicaciones en la nube sin necesidad de intervención por
parte del usuario.
Evaluación de hipótesis y análisis de resultados. Tras validar las hipótesis propuestas,
se analizan los resultados para extraer las consiguientes conclusiones de toda la
investigación realizada.
Documentación. La tarea de documentación se ha llevado a cabo a lo largo de todo el
proceso de elaboración de la presente tesis doctoral. En las primeras fases, la tarea de
documentación registra los resultados del estudio del estado del arte y del análisis del
problema. Durante el desarrollo de las siguientes etapas, se documentaron las
características del diseño realizado y los aspectos relevantes del sistema global. En la
fase de validación, se han documentado los resultados de los procedimientos de
validación seguidos. Finalmente se redactó la documentación definitiva que constituye
la presente memoria, incluyendo las conclusiones extraídas de cada uno de los
aspectos relativos a la tesis doctoral, a partir de la documentación generada a lo largo
de las distintas fases.
La evaluación del lenguaje y del sistema global en el que se enmarca, ha probado la viabilidad
de la solución presentada para alcanzar la interoperabilidad entre aplicaciones heterogéneas a
nivel de datos en entornos de computación en nube de manera automática. Gracias a la acción
conjunta de todos los componentes integrantes de la arquitectura diseñada se han conseguido
resultados satisfactorios en procesos de intercambio de información entre aplicaciones
situadas en una misma estructura de Cloud Computing. El proceso de validación ha permitido
además demostrar cada una de las hipótesis planteadas por la investigación. Por todo ello es
posible afirmar que se han alcanzado los objetivos planteados por la tesis doctoral, resultando
finalmente en un gran éxito y una contribución de relevancia en los diferentes ámbitos
estudiados. Además, las conclusiones extraídas de la totalidad de la investigación permiten
extrapolar aspectos teóricos a otros medios y abrir nuevas líneas de investigación destinadas a
contribuir a la consecución de entornos interoperables que permitan un mejor y más eficaz
aprovechamiento de la información existente. ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
The Internet of services has altered the Web from a simple repository of information to become a services and transactions platform where organizations offer different kinds of services, increasing their business processes through the Web. With this transformation, new paradigm mergers have occurred, such as Cloud Computing and Software-as-a-Service, whose models are adapted to this new conception and promise the creation of new levels of efficiency through large scale sharing of functionalities and resources. With all of this, Cloud Computing related technologies have increased their presence and importance in the world of information technologies and have evolved to become a mature set of technological innovations able to provide a strong infrastructure for the SaaS paradigm. Cloud Computing and Software-as-a-Service open the doors to large scale economies and new horizons, but they have to face an important number of challenges, among which we can find the following: The lack of trusted models in order to determine the conditions under which it would be worthwhile for the organization to start a process of migration to these kinds of models. The lack of tested methods (for example, architecture guides) that would facilitate a potential migration. The lack of integration methods at different levels. This complicates the development and delivery processes of software that is able to communicate with other elements in the cloud platform, therefore causing to lose one of its most important features. In dealing with this last aspect, it is possible to find the concept of interoperability. Interoperability has always been considered as one of the major challenges in information systems given that the heterogeneity of systems raises the complexity of the necessary methods to achieve it. This problem is growing with the increasing proliferation of independent-built information systems, which obstruct or impede the information exchange between them. Due to this, one of the most interesting properties of Cloud Computing systems is missing and, furthermore, it is restricting the progress towards the possibility of a completely interoperable world where the access to information would be nearly infinite. The problem of interoperability has been faced from different perspectives, however the ones based on semantic technologies, where reaching interoperability is historically one of their major goals, are especially interesting. The Semantic Web was conceived for in order to solve the problems of data overloading and heterogeneity in the Web, which causes the absence of interoperability among other features. The approach of Semantic Web is to add semantic to the data enabling the machine´s capability to process, reason, combine and make logical deductions with them in a similar way than a human will do, providing a set of useful tools. Within this field, this doctoral thesis brings together the current most relevant and rising concepts like Cloud Computing, SaaS, semantic technologies, business process modeling, systems interoperability, etc. The goal is to encourage the evolution of new platforms oriented towards interoperability and cost reduction, which can have a significant impact in the industry. The research accomplished takes these technologies and paradigms as a base for the definition of a standard or common language that allows the interoperability between applications hosted in the same Cloud Computing environment. With this, it is possible to take a big step forward in the technology areas previously mentioned, creating a clearly innovative approach with results that will obtain great significance in the world of information systems. With the information mentioned, the solution framed within this doctoral thesis is oriented towards the design of a semantic language ontology-based and of a global system that works around it, able to capture the knowledge of the applications hosted in a Cloud Computing structure, setting up the development of information exchange processes between heterogeneous applications in an automatic manner. Therefore, this thesis proposes the combination of Cloud Computing and Semantics in order to tackle the problem of interoperability at application level. Currently this approach is not widespread due to the “recent” explosion of cloud systems, but the benefits that they can mutually provide translate into a great motivation for working and deeply researching in this field. The research methodology followed during this doctoral thesis in order to achieve the goals and provide the formulated hypothesis consists of: State of the art study. Through the study of related works, it is possible to know the methods and technologies necessary to achieve the outlined goals. This study is divided in five main categories: o Cloud Computing. o Semantic Web. o Web Services. o Semantic Web Services. o Interoperability. Analysis of the Literature’s most representative aspects. The analysis of the different concepts addressed in the state of the art, allows performing a correct and strongly supported decision process to utilize the most appropriate technologies and tools for the problem being solved. First approach of the solution. This stage allows for the layout of the foundations and requirements that have to be fulfilled in order to reach the proposed objectives. This researching phase makes use of the conclusions extracted from the previous analysis and allows to determine the viability of the researching as well as checking whether the selected tools are good enough to achieve the desired solution. Final design of the solution. In this phase the design process concludes. The whole set of requirements have to be fulfilled and it must be detailed enough to be able to start the implementation of the environment for its validation. Language implementation and configuration of Cloud Computing platform. This stage indicates the beginning of the validation process which will allow the testing of whether the design meets the objectives and is aligned with the proposed hypothesis. The configuration of the Cloud Computing platform will enable evaluating and validating the researching in a real environment. Validation. The whole designed and developed functionality is tested within this phase in order to ensure that the problem is correctly represented and the system is able to perform the different tasks to accomplish automatic interoperability. Hypothesis evaluation and analysis of results. After validating the proposed hypothesis, results are analyzed to extract the necessary conclusions related to the implemented research. Documentation. This task has been developed throughout the whole process of this doctoral thesis preparation. In the first stages, the documentation registers the results obtained from the state of the art study and the analysis of the problem. During the following phases, all the features of the developed design and the relevant aspects from the global system were documented. Within the validation phase, the results of the performed procedures have been written. Lastly, the definitive statement that makes this document was composed, including the conclusions extracted from each of the aspects relative to this doctoral thesis, arising from the documentation generated along the different phases. The evaluation of the language, and the global system where it is framed, has proved the viability of the proposed solution, achieving automatic interoperability of heterogeneous applications at data level in Cloud Computing environments. Thanks to the joint action of all components forming the designed architecture, this research provided satisfactory results in the information exchange processes between applications hosted in the same Cloud Computing infrastructure. Moreover, the validation process has allowed for the demonstration of each of the hypotheses proposed by the research. By the information contained herein, it is possible to affirm that all the objectives exposed in this doctoral thesis have been accomplished, resulting in a great success and a relevant contribution to the different areas analyzed. Furthermore, the conclusions extracted from the whole research allow for the extrapolation of some theoretical aspects to other environments, as well as open new research lines that contribute to the achievement of interoperable environments that are able to provide a better and more efficient use of the vast volume of existing information.
The Internet of services has altered the Web from a simple repository of information to become a services and transactions platform where organizations offer different kinds of services, increasing their business processes through the Web. With this transformation, new paradigm mergers have occurred, such as Cloud Computing and Software-as-a-Service, whose models are adapted to this new conception and promise the creation of new levels of efficiency through large scale sharing of functionalities and resources. With all of this, Cloud Computing related technologies have increased their presence and importance in the world of information technologies and have evolved to become a mature set of technological innovations able to provide a strong infrastructure for the SaaS paradigm. Cloud Computing and Software-as-a-Service open the doors to large scale economies and new horizons, but they have to face an important number of challenges, among which we can find the following: The lack of trusted models in order to determine the conditions under which it would be worthwhile for the organization to start a process of migration to these kinds of models. The lack of tested methods (for example, architecture guides) that would facilitate a potential migration. The lack of integration methods at different levels. This complicates the development and delivery processes of software that is able to communicate with other elements in the cloud platform, therefore causing to lose one of its most important features. In dealing with this last aspect, it is possible to find the concept of interoperability. Interoperability has always been considered as one of the major challenges in information systems given that the heterogeneity of systems raises the complexity of the necessary methods to achieve it. This problem is growing with the increasing proliferation of independent-built information systems, which obstruct or impede the information exchange between them. Due to this, one of the most interesting properties of Cloud Computing systems is missing and, furthermore, it is restricting the progress towards the possibility of a completely interoperable world where the access to information would be nearly infinite. The problem of interoperability has been faced from different perspectives, however the ones based on semantic technologies, where reaching interoperability is historically one of their major goals, are especially interesting. The Semantic Web was conceived for in order to solve the problems of data overloading and heterogeneity in the Web, which causes the absence of interoperability among other features. The approach of Semantic Web is to add semantic to the data enabling the machine´s capability to process, reason, combine and make logical deductions with them in a similar way than a human will do, providing a set of useful tools. Within this field, this doctoral thesis brings together the current most relevant and rising concepts like Cloud Computing, SaaS, semantic technologies, business process modeling, systems interoperability, etc. The goal is to encourage the evolution of new platforms oriented towards interoperability and cost reduction, which can have a significant impact in the industry. The research accomplished takes these technologies and paradigms as a base for the definition of a standard or common language that allows the interoperability between applications hosted in the same Cloud Computing environment. With this, it is possible to take a big step forward in the technology areas previously mentioned, creating a clearly innovative approach with results that will obtain great significance in the world of information systems. With the information mentioned, the solution framed within this doctoral thesis is oriented towards the design of a semantic language ontology-based and of a global system that works around it, able to capture the knowledge of the applications hosted in a Cloud Computing structure, setting up the development of information exchange processes between heterogeneous applications in an automatic manner. Therefore, this thesis proposes the combination of Cloud Computing and Semantics in order to tackle the problem of interoperability at application level. Currently this approach is not widespread due to the “recent” explosion of cloud systems, but the benefits that they can mutually provide translate into a great motivation for working and deeply researching in this field. The research methodology followed during this doctoral thesis in order to achieve the goals and provide the formulated hypothesis consists of: State of the art study. Through the study of related works, it is possible to know the methods and technologies necessary to achieve the outlined goals. This study is divided in five main categories: o Cloud Computing. o Semantic Web. o Web Services. o Semantic Web Services. o Interoperability. Analysis of the Literature’s most representative aspects. The analysis of the different concepts addressed in the state of the art, allows performing a correct and strongly supported decision process to utilize the most appropriate technologies and tools for the problem being solved. First approach of the solution. This stage allows for the layout of the foundations and requirements that have to be fulfilled in order to reach the proposed objectives. This researching phase makes use of the conclusions extracted from the previous analysis and allows to determine the viability of the researching as well as checking whether the selected tools are good enough to achieve the desired solution. Final design of the solution. In this phase the design process concludes. The whole set of requirements have to be fulfilled and it must be detailed enough to be able to start the implementation of the environment for its validation. Language implementation and configuration of Cloud Computing platform. This stage indicates the beginning of the validation process which will allow the testing of whether the design meets the objectives and is aligned with the proposed hypothesis. The configuration of the Cloud Computing platform will enable evaluating and validating the researching in a real environment. Validation. The whole designed and developed functionality is tested within this phase in order to ensure that the problem is correctly represented and the system is able to perform the different tasks to accomplish automatic interoperability. Hypothesis evaluation and analysis of results. After validating the proposed hypothesis, results are analyzed to extract the necessary conclusions related to the implemented research. Documentation. This task has been developed throughout the whole process of this doctoral thesis preparation. In the first stages, the documentation registers the results obtained from the state of the art study and the analysis of the problem. During the following phases, all the features of the developed design and the relevant aspects from the global system were documented. Within the validation phase, the results of the performed procedures have been written. Lastly, the definitive statement that makes this document was composed, including the conclusions extracted from each of the aspects relative to this doctoral thesis, arising from the documentation generated along the different phases. The evaluation of the language, and the global system where it is framed, has proved the viability of the proposed solution, achieving automatic interoperability of heterogeneous applications at data level in Cloud Computing environments. Thanks to the joint action of all components forming the designed architecture, this research provided satisfactory results in the information exchange processes between applications hosted in the same Cloud Computing infrastructure. Moreover, the validation process has allowed for the demonstration of each of the hypotheses proposed by the research. By the information contained herein, it is possible to affirm that all the objectives exposed in this doctoral thesis have been accomplished, resulting in a great success and a relevant contribution to the different areas analyzed. Furthermore, the conclusions extracted from the whole research allow for the extrapolation of some theoretical aspects to other environments, as well as open new research lines that contribute to the achievement of interoperable environments that are able to provide a better and more efficient use of the vast volume of existing information.
Description
Keywords
Servicios Web, Tecnologías de la información, Cloud computing, Computación en nube, Ingeniería del conocimiento