|Bibliography||Pincay Nieves, Jhonny Vladimir: Knowledge capturing and usage of evolving cloud application topologies. |
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 67 (2016).
137 pages, english.
In the last few years, the advent of Cloud Computing has contributed to a considerable increase of the number of service offerings from a variety of providers and every time more applications are being partially or fully deployed in the cloud. Nowadays, experts invest considerable time and effort towards taking the maximum advantage of the benefits that the employment of cloud technologies brings. Such a wide spectrum of cloud offerings, however, increase the number of complex tasks and aspects that must be taken into account when distributing the components of an application such as the quality of service, adaptation to market changes, etc. Furthermore, the distribution of the components and the configuration of cloud resources may evolve over time and there is a lack of tooling support when it comes to assist the developers in the decision-making tasks of selecting an appropriate application distribution topology. One of the methods that human-beings use to solve problems consists of recalling past experiences and how they solved them at that time. Under this and other considerations the Case-Based Reasoning paradigm was conceived, as a mechanism for solving tasks by recalling past similar problems and adapting their solutions to new situations. This work aims to develop the concepts and mechanisms that enable the capturing and usage of knowledge that the evolution of a system brings along. Specifically, this thesis attempts to identify a set of characteristics that accurately describe a cloud application and to define the models that should be used to identify the cases solved in the past and whose solutions may be useful to solve a new problem. To achieve this, the Case-Based Reasoning with Similarity Retrieval approach is employed, to identify applications with similar characteristics, retrieve their solutions and offer means to refine them in order to obtain a distribution topology that fulfills the requirements of a given application. Furthermore, a prototypical implementation of the approach of this thesis is executed and also employed to validate the concepts and principles that this work follows.
|Full text and|
|Department(s)||University of Stuttgart, Institute of Architecture of Application Systems|
|Superviser(s)||Leymann, Prof. Frank; Gómez Sáez, Santiago|
|Entry date||June 5, 2019|