1,720,965 research outputs found
Physical Database Design in Document Stores
No full textNoSQL is an umbrella term used to classify alternate storage systems to the traditional Relational Database Management Systems (RDBMSs). At the moment of writing, there are more than 200 NoSQL systems available that can be classified into four main categories on the data storage model: key-value stores, document stores, column family stores, and graph stores. Document stores have gained popularity mainly due to the semi-structured data storage model and the rich query capabilities compared to the other NoSQL systems making them an ideal candidate for rapid prototyping. Document stores encourage users to use a data-first approach as opposed to a design-first one. Database design on document stores is mainly carried out in a trial-and-error or ad-hoc rule-based manner instead of a formal process such as normalization in an RDBMS. However, these approaches could easily lead to a non-optimal database design leading to additional costs in query processing, data storage, and redesigning.This PhD thesis aims to provide a novel multi-criteria-based approach to database design in document stores. Most of the existing approaches of database design are based on optimizing query performance. However, other factors include storage requirement and complexity of the stored documents specific to each use case. Moreover, there is a large solution space of alternative designs due to the different combinations of referencing and nesting of data. Hence, we believe multi-criteria optimization is ideal with a proven track record of solving such problems in various domains. However, to achieve this, we need to address several issues that will enable us to apply multi-criteria optimization for the data design problem.First, we evaluate the impact of alternate storage representations of semi-structured data. There are multiple and equivalent ways to physically represent semi-structured data, but there is a lack of evidence about the potential impact on space and query performance. Thus, we embark on the task of quantifying that precisely for document stores. We empirically compare multiple ways of representing semi-structured data, which allows us to derive a set of guidelines for efficient physical database design considering both JSON and relational options in the same palette.Then, we need a formal canonical model that is capable of representing alternative designs. To this extent, we propose a hypergraph-based approach for representing heterogeneous datastore designs. Taking an existing common programming interface to NoSQL systems, we extend and formalize it as hypergraphs. Then, we define design constraints and query transformation rules for three representative data store types. Next, we propose a simple query rewriting algorithm from a generic one into underlying data stores specific one and provide a prototype implementation. Furthermore, we introduce a storage statistics estimator on the underlying data stores. Finally, we show the feasibility of our approach on a use case of an existing polyglot system and its usefulness in metadata and physical query path calculations.Next, we require a formal query cost model to estimate and evaluate query performance on alternative document store designs. Document stores use primitive approaches to query processing, such as evaluating all possible query plans to find the winning one and using it in the subsequent similar queries or relying on the end-user to specify the usage of indexes instead of a formal cost model. However, we require a reliable approach to compare two alternative designs on how they perform on a specific query. For this, we define a generic storage and query cost model based on disk access and memory allocation that allows estimating the impact of design decisions. Since all document stores carry out data operations in memory, we first estimate the memory usage by considering the characteristics of the stored documents, their access patterns, and memory management algorithms. Then, using this estimation and metadata storage size, we introduce a cost model for random access queries. This is the first attempt at such an approach to the best of our knowledge. Finally, we validate our work on two well-known document store implementations: MongoDB and Couchbase. The results show that the memory usage estimates have an average precision of 91%, and predicted costs are highly correlated to the actual execution times. During this work, we have managed to suggest several improvements to document storage systems. Thus, this cost model also contributes to identifying discordance between document store implementations and their theoretical expectations.Finally, we implement the automated database design solution using multi-criteria optimization. First, we introduce an algebra of transformations that can systematically modify a design of our canonical representation. Then, using these transformations, we implement a local search algorithm driven by a loss function that can propose near-optimal designs with high probability. Finally, we compare our prototype against an existing document store data design solution purely driven by query cost. Our proposed designs have better performance and are more compact with less redundancy.Doctorat en Sciences de l'ingénieur et technologieinfo:eu-repo/semantics/nonPublishe
Physical database design in document stores
Get PDFTesi en modalitat de cotutela, Universitat Politècnica de Catalunya i Université libre de BruxellesNoSQL is an umbrella term used to classify alternate storage systems to the traditional Relational Database Management Systems (RDBMSs). Among these, Document stores have gained popularity mainly due to the semi-structured data storage model and the rich query capabilities. They encourage users to use a data-first approach as opposed to a design-first one. Database design on document stores is mainly carried out in a trial-and-error or ad-hoc rule-based manner instead of a formal process such as normalization in an RDBMS. However, these approaches could easily lead to a non-optimal design resulting additional costs in the long run.
This PhD thesis aims to provide a novel multi-criteria-based approach to database design in document stores. Most of such existing approaches are based on optimizing query performance. However, other factors include storage requirement and complexity of the stored documents specific to each use case. There is a large solution space of alternative designs due to the different combinations of referencing and nesting of data. Thus, we believe multi-criteria optimization is ideal to solve this problem. To achieve this, we need to address several issues that will enable us to apply multi-criteria optimization for the data design problem.
First, we evaluate the impact of alternate storage representations of semi-structured data. There are multiple and equivalent ways to physically represent semi-structured data, but there is a lack of evidence about the potential impact on space and query performance. Thus, we embark on the task of quantifying that precisely for document stores. We empirically compare multiple ways of representing semi-structured data, allowing us to derive a set of guidelines for efficient physical database design considering both JSON and relational options in the same palette.
Then, we need a formal canonical model that can represent alternative designs. We propose a hypergraph-based approach for representing heterogeneous datastore designs. We extend and formalize an existing common programming interface to NoSQL systems as hypergraphs. We define design constraints and query transformation rules for representative data store types. Next, we propose a simple query rewriting algorithm and provide a prototype implementation together with storage statistics estimator.
Next, we require a formal query cost model to estimate and evaluate query performance on alternative document store designs.
Document stores use primitive approaches to query processing, such as relying on the end-user to specify the usage of indexes instead of a formal cost model. But we require a reliable approach to compare alternative designs on how they perform on a specific query. For this, we define a generic storage and query cost model based on disk access and memory allocation. As all document stores carry out data operations in memory, we first estimate the memory usage by considering the characteristics of the stored documents, their access patterns, and memory management algorithms. Then, using this estimation and metadata storage size, we introduce a cost model for random access queries. We validate our work on two well-known document store implementations.
The results show that the memory usage estimates have an average precision of 91% and predicted costs are highly correlated to the actual execution times. During this work, we also managed to suggest several improvements to document stores.
Finally, we implement the automated database design solution using multi-criteria optimization. We introduce an algebra of transformations that can systematically modify a design of our canonical representation. Then, using them, we implement a local search algorithm driven by a loss function that can propose near-optimal designs with high probability. We compare our prototype against an existing document store data design solution. Our proposed designs have better performance and are more compact with less redundancy.NoSQL descriu sistemes d'emmagatzematge alternatius als tradicionals de gestió de bases de dades relacionals (RDBMS). Entre aquests, els magatzems de documents han guanyat popularitat principalment a causa del model de dades semiestructurat i les riques capacitats de consulta. Animen els usuaris a utilitzar un enfocament de dades primer, en lloc d'un enfocament de disseny primer. El disseny de dades en magatzems de documents es porta a terme principalment en forma d'assaig-error o basat en regles ad-hoc en lloc d'un procés formal i sistemàtic com ara la normalització en un RDBMS. Aquest enfocament condueix fàcilment a un disseny no òptim que generarà costos addicionals a llarg termini. La majoria dels enfocaments existents es basen en l'optimització del rendiment de les consultes. Aquesta tesi pretén, en canvi, proporcionar un nou enfocament basat en diversos criteris per al disseny de bases de dades en magatzems de documents, inclouen el requisit d'espai i la complexitat dels documents emmagatzemats específics per a cada cas d'ús. En general, hi ha un gran espai de solucions de dissenys alternatives. Per tant, creiem que l'optimització multicriteri és ideal per resoldre aquest problema. Per aconseguir-ho, hem d'abordar diversos problemes que ens permetran aplicar l'optimització multicriteri. En primer, avaluem l'impacte de les representacions alternatives de dades semiestructurades. Hi ha maneres múltiples i equivalents de representar dades semiestructurades, però hi ha una manca d'evidència sobre l'impacte potencial en l'espai i el rendiment de les consultes. Així, ens embarquem en la tasca de quantificar-ho. Comparem empíricament múltiples representacions de dades semiestructurades, cosa que ens permet derivar directrius per a un disseny eficient tenint en compte les opcions dels JSON i relacionals alhora. Aleshores, necessitem un model canònic que pugui representar dissenys alternatius i proposem un enfocament basat en hipergrafs. Estenem i formalitzem una interfície de programació comuna existent als sistemes NoSQL com a hipergrafs. Definim restriccions de disseny i regles de transformació de consultes per a tipus de magatzem de dades representatius. A continuació, proposem un algorisme de reescriptura de consultes senzill i proporcionem una implementació juntament amb un estimador d'estadístiques d'emmagatzematge. Els magatzems de documents utilitzen enfocaments primitius per al processament de consultes, com ara confiar en l'usuari final per especificar l'ús d'índexs en lloc d'un model de cost. Conseqüentment, necessitem un model de cost de consulta per estimar i avaluar el rendiment en dissenys alternatius. Per això, definim un model genèric propi basat en l'accés a disc i l'assignació de memòria. Com que tots els magatzems de documents duen a terme operacions de dades a memòria, primer estimem l'ús de la memòria tenint en compte les característiques dels documents emmagatzemats, els seus patrons d'accés i els algorismes de gestió de memòria. A continuació, utilitzant aquesta estimació i la mida d'emmagatzematge de metadades, introduïm un model de costos per a consultes d'accés aleatori. Validem el nostre treball en dues implementacions conegudes. Els resultats mostren que les estimacions d'ús de memòria tenen una precisió mitjana del 91% i els costos previstos estan altament correlacionats amb els temps d'execució reals. Finalment, implementem la solució de disseny automatitzat de bases de dades mitjançant l'optimització multicriteri. Introduïm una àlgebra de transformacions que pot modificar sistemàticament un disseny en la nostra representació canònica. A continuació, utilitzant-la, implementem un algorisme de cerca local impulsat per una funció de pèrdua que pot proposar dissenys gairebé òptims amb alta probabilitat. Comparem el nostre prototip amb una solució de disseny de dades de magatzem de documents existent. Els nostres dissenys proposats tenen un millor rendiment i són més compactes, amb menys redundànciaNoSQL est un terme générique utilisé pour classer les systèmes de stockage alternatifs aux systèmes de gestion de bases de données relationnelles (SGBDR) traditionnels. Au moment de la rédaction de cet article, il existe plus de 200 systèmes NoSQL disponibles qui peuvent être classés en quatre catégories principales sur le modèle de stockage de données : magasins de valeurs-clés, magasins de documents, magasins de familles de colonnes et magasins de graphiques. Les magasins de documents ont gagné en popularité principalement en raison du modèle de stockage de données semi-structuré et des capacités de requêtes riches par rapport aux autres systèmes NoSQL, ce qui en fait un candidat idéal pour le prototypage rapide. Les magasins de documents encouragent les utilisateurs à utiliser une approche axée sur les données plutôt que sur la conception. La conception de bases de données sur les magasins de documents est principalement effectuée par essais et erreurs ou selon des règles ad hoc plutôt que par un processus formel tel que la normalisation dans un SGBDR. Cependant, ces approches pourraient facilement conduire à une conception de base de données non optimale entraînant des coûts supplémentaires de traitement des requêtes, de stockage des données et de refonte.
Cette thèse de doctorat vise à fournir une nouvelle approche multicritère de la conception de bases de données dans les magasins de documents.
La plupart des approches existantes de conception de bases de données sont basées sur l’optimisation des performances des requêtes. Cependant, d’autres facteurs incluent les exigences de stockage et la complexité des documents stockés spécifique à chaque cas d’utilisation. De plus, il existe un grand espace de solution de conceptions alternatives en raison des différentes combinaisons de référencement et d’imbrication des données. Par conséquent, nous pensons que l’optimisation multicritères est idéale par l’intermédiaire d’une expérience éprouvée dans la résolution de tels problèmes dans divers domaines. Cependant, pour y parvenir, nous devons résoudre plusieurs problèmes qui nous permettront d’appliquer une optimisation multicritère pour le problème de conception de données.
Premièrement, nous évaluons l’impact des représentations alternatives de stockage des données semi-structurées. Il existe plusieurs manières équivalentes de représenter physiquement des données semi-structurées, mais il y a un manque de preuves concernant l’impact potentiel sur l’espace et sur les performances des requêtes. Ainsi, nous nous lançons dans la tâche de quantifier cela précisément pour les magasins de documents. Nous comparons empiriquement plusieurs façons de représenter des données semi-structurées, ce qui nous permet de dériver un ensemble de directives pour une conception de base de données physique efficace en tenant compte à la fois des options JSON et relationnelles dans la même palette.
Ensuite, nous avons besoin d’un modèle canonique formel capable de représenter des conceptions alternatives. Dans cette mesure, nous proposons une approche basée sur des hypergraphes pour représenter des conceptions de magasins de données hétérogènes. Prenant une interface de programmation commune existante aux systèmes NoSQL, nous l’étendons et la formalisons sous forme d’hypergraphes. Ensuite, nous définissons les contraintes de conception et les règles de transformation des requêtes pour trois types de magasins de données représentatifs. Ensuite, nous proposons un algorithme de réécriture de requête simple à partir d’un algorithme générique dans un magasin de données sous-jacent spécifique et fournissons une implémentation prototype. De plus, nous introduisons un estimateur de statistiques de stockage sur les magasins de données sous-jacents. Enfin, nous montrons la faisabilité de notre approche sur un cas d’utilisation d’un système polyglotte existant ainsi que son utilité dans les calculs de métadonnées et de chemins de requêtes physiques.
Ensuite, nous avons besoin d’un modèle de coûts de requêtes formel pour estimer et évaluer les performances des requêtes sur des conceptions alternatives de magasin de documents. Les magasins de documents utilisent des approches primitives du traitement des requêtes, telles que l’évaluation de tous les plans de requête possibles pour trouver le plan gagnant et son utilisation dans les requêtes similaires ultérieures, ou l’appui sur l’usager final pour spécifier l’utilisation des index au lieu d’un modèle de coûts formel.
Cependant, nous avons besoin d’une approche fiable pour comparer deux conceptions alternatives sur la façon dont elles fonctionnent sur une requête spécifique. Pour cela, nous définissons un modèle de coûts de stockage et de requête générique basé sur l’accès au disque et l’allocation de mémoire qui permet d’estimer l’impact des décisions de conception. Étant donné que tous les magasins de documents effectuent des opérations sur les données en mémoire, nous estimons d’abord l’utilisation de la mémoire en considérant les caractéristiques des documents stockés, leurs modèles d’accès et les algorithmes de gestion de la mémoire. Ensuite, en utilisant cette estimation et la taille de stockage des métadonnées, nous introduisons un modèle de coûts pour les requêtes à accès aléatoire. Il s’agit de la première tenta ive d’une telle approche au meilleur de notre connaissance. Enfin, nous validons notre travail sur deux implémentations de magasin de documents bien connues : MongoDB et Couchbase. Les résultats démontrent que les estimations d’utilisation de la mémoire ont une précision moyenne de 91% et que les coûts prévus sont fortement corrélés aux temps d’exécution réels. Au cours de ce travail, nous avons réussi à proposer plusieurs améliorations aux systèmes de stockage de documents. Ainsi, ce modèle de coûts contribue également à identifier les discordances entre les implémentations de stockage de documents et leurs attentes théoriques.
Enfin, nous implémentons la solution de conception automatisée de bases de données en utilisant l’optimisation multicritères. Tout d’abord, nous introduisons une algèbre de transformations qui peut systématiquement modifier une conception de notre représentation canonique. Ensuite, en utilisant ces transformations, nous implémentons un algorithme de recherche locale piloté par une fonction de perte qui peut proposer des conceptions quasi optimales avec une probabilité élevée. Enfin, nous comparons notre prototype à une solution de conception de données de magasin de documents existante uniquement basée sur le coût des requêtes. Nos conceptions proposées ont de meilleures performances et sont plus compactes avec moins de redondanceDoctorat Erasmus Mundus en tecnologies de la informació per a la intel·ligència empresarialDOCTORAT ERASMUS MUNDUS EN TECNOLOGIES DE LA INFORMACIÓ PER A LA INTEL·LIGÈNCIA EMPRESARIAL (Pla 2013
A cost model for random access queries in document stores
Get PDFDocument stores have become one of the key NoSQL storage solutions. They have been widely adopted in different domains due to their ability to store semi-structured data and expressive query capabilities. However, implementations differ in terms of concrete data storage and retrieval. Unfortunately, a standard framework for data and query optimization for document stores is nonexistent, and only implementation-specific design and query guidelines are used. Hence, the goal of this work is to aid automating the data design for document stores based on query costs instead of generic design rules. For this, we define a generic storage and query cost model based on disk access and memory allocation that allows estimating the impact of design decisions. Since all document stores carry out data operations in memory, we first estimate the memory usage by considering characteristics of the stored documents, their access patterns, and memory management algorithms. Then, using this estimation and metadata storage size, we introduce a cost model for random access queries. We validate our work on two well-known document store implementations: MongoDB and Couchbase. The results show that the memory usage estimates have the average precision of 91% and predicted costs are highly correlated to the actual execution times. During this work, we have managed to suggest several improvements to document storage systems. Thus, this cost model also contributes to identifying discordance between document store implementations and their theoretical expectations.This research has been funded by the European Commission through the Erasmus Mundus Joint Doctorate Information Technologies for Business Intelligence- Doctoral College (IT4BI-DC)Peer ReviewedPostprint (author's final draft
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Get PDFWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
