In this thesis I detail an easy to use system for dynamic data storage aimed at use in educational games and game systems. Evaluation of current approaches in use by educators and students suggest a large degree of ad hoc systems in place for data persistence. I evaluate my system against existing techniques including a flat file, XML, and SQL systems. I provide a dynamic, easy to use solution to data persistence in the environment of educational game development.

As data and data structures grown more complex in computing, the task of storing and accessing such data efficiently also becomes more complex. Object-oriented programming languages such as Java have popularized the practice of using class-based objects and created the challenge of persisting these complex objects to storage systems such as databases which only store simple scalar values. This paper seeks to explore and compare a selected num-ber of popular persistence solutions for the Java language in their use and per-formance. This is done through researching, implementing and performance testing the chosen candidates. Through these steps we have found that while more abstracted solutions provided easier implementation and usage, these positives come with the disadvantages of generally slower performance. We also discovered that while many persistence frameworks provide features that minimize the amount of code required for implementation, they suffer from performance issues such as the N+1 query issue if not configured or utilized correctly.

Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2017-01-16T10:13:02Z No. of bitstreams: 2 Dissertação - Áurea Valéria Pereira da Silva - 2016.pdf: 2576793 bytes, checksum: 3b3472812df3319818244bcd2e0482d7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-01-16T10:13:30Z (GMT) No. of bitstreams: 2 Dissertação - Áurea Valéria Pereira da Silva - 2016.pdf: 2576793 bytes, checksum: 3b3472812df3319818244bcd2e0482d7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-01-16T10:13:30Z (GMT). No. of bitstreams: 2 Dissertação - Áurea Valéria Pereira da Silva - 2016.pdf: 2576793 bytes, checksum: 3b3472812df3319818244bcd2e0482d7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-12-14 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Introduction: OpenEHR is a standardization of Health Information Systems (HIS) that is mainly concerned with the exchange of Electronic Health Records (EHR). One of the major obstacles to the adoption of openEHR is the lack of database persistence specifications. Objective: The aim of this work is the mapping of storage structures from the Archetype Definition Language (ADL) specifications which describe clinical knowledge structure. Method: This research initially makes a systematic mapping about persistence structures that are derived directly from ADL specifications, i.e. without dependences from the Reference Model (RM) of openEHR. A new persistence strategy is proposed and compared with ones found in the literature. Results: Assumptions, criteria and rules were used to present the proposed approach. Its evaluation considered quantitative aspects of persistence structures. Conclusion: The evaluation revealed some advantages of proposed approach, such as: reductions of reference attributes (referential integrity) and redundancy in clinical database; production of unidentified clinical records in relation to the patient; creation of a single base table for each archetype, regardless of its use in the form of slots; scalability of database schema (stable number of database tables), even in the occurrence of recursion through slots. / Introdução: OpenEHR é uma padronização dos Sistemas de Informação em Saúde (SIS) que se preocupa principalmente com a troca de Registros de Saúde Eletrônicos (RES). Um dos maiores obstáculos à adoção do openEHR é a carência de especificações de persistência de banco de dados. Objetivo: O objetivo deste trabalho é o mapeamento de estruturas de armazenamento a partir das especificações Archetype Definition Language (ADL) que descrevem a estrutura do conhecimento clínico. Métodos: Esta pesquisa inicialmente faz um mapeamento sistemático sobre estruturas de persistência que são derivadas diretamente de especificações ADL, isto é, sem dependências do Modelo de Referência (RM) de openEHR. Uma nova estratégia de persistência é proposta e comparada com as encontradas na literatura. Resultados: Foram utilizados pressupostos, critérios e regras para apresentar a abordagem proposta. Uma avaliação considerou aspectos quantitativos das estruturas de persistência, em comparação com o que foi encontrado na literatura. Conclusões: A avaliação revelou algumas vantagens da abordagem proposta, tais como: reduções de atributos de referência (integridade referencial) e redundância em banco de dados clínicos; produção de registros clínicos não identificados em relação ao paciente; criação de uma tabela de base única para cada arquétipo, independentemente da sua utilização sob a forma de slots; escalabilidade do esquema de banco de dados (número estável de tabelas de banco de dados), mesmo na ocorrência de recursão através de slots.