databases_and_information_systems.md

Notes About Databases and Information Systems

Table of Contents

• Definitions of Database, Database Management System, and Related Concepts
• Ancillary Definitions
  • Side Notes on Definitions
• Database Design
• Database Models
• Categories of Databases
  • Object Databases
  • SQL
  • NoSQL
    • Key-Value Store
    • NoSQL Database Management Systems
  • NewSQL
• Distributed Databases for Big Data
  • Distributed Databases Based on MapReduce
• Database Data Formats

To-Do List: Tasks

References

Author Information

Definitions of Database, Database Management System, and Related Concepts

A database is a collection of data that is systematically organized for access, storage, and processing \cite{DictionaryDotComStaff2018,WikipediaContributors2018r}.

A "general-purpose" database management system (DBMS) is a software system that enables people to create, maintain, and access databases \cite{WikipediaContributors2018r}.

The main functions of DBMS are \cite{WikipediaContributors2018r}:

• data definition, data is organized according to definitions of relationships between the data; these definitions need to be created, modified, and deleted
• update, which allows data to be added, modified, or removed from the database
• retrieval, which allows data to be accessed
• administration, which involves registering people to access the database and monitoring them, and implementing techniques for data security and data integrity, performance monitoring, concurrency control, and data/information recovery

A database model is an abstract model (specifically, a data model) that describes the logical structure of a database, and how to store, organize, and manipulate data \cite{WikipediaContributors2018w}.

A data model (or datamodel) is a description of how data is organized and related to each other, and the relationships between data and entities in the real world \cite{WikipediaContributors2018x}; We can create data models using entities, attributes, relations, and tables \cite{WikipediaContributors2018x}. In enterprise modeling, a function model complements a data model \cite{WikipediaContributors2018x}. The main categories of data model instances are \cite{WikipediaContributors2018x}:

• conceptual data model, which is a technology-independent description of the semantics of a specific domain (in the information context as abstract structures \cite{WikipediaContributors2018y}) and defines the scope of the model
  • "Includes high-level data constructs" that are used to create architectural descriptions in layperson terms \cite{WikipediaContributors2018y}.
  • Avoids technical names to facilitate understanding of the architectural description's data basis \cite{WikipediaContributors2018y}.
  • "May not be normalized" \cite{WikipediaContributors2018y}.
• logical data model (or logical schema \cite{WikipediaContributors2018y}), which describes the semantics of a conceptual model for "a particular data manipulation technology", in terms of (relational) tables, columns, object-oriented classes, and XML tags
  • A logical data model is a detailed reflection of a conceptual data model's domain-specific semantics, and is independent of the physical data model ("particular database management product or storage technology" \cite{WikipediaContributors2018y})
  • It is a representation of a domain-specific abstract structure (or, they "represent the abstract structure of a domain of information") \cite{WikipediaContributors2018y};
  • Use the logical data model to create databases, and use it as the basis of a physical data model \cite{WikipediaContributors2018y}
  • types of logical data models \cite{WikipediaContributors2018y}
    • hierarchical data model \cite{WikipediaContributors2018w}.
    • network data model \cite{WikipediaContributors2018w}
    • relational model \cite{WikipediaContributors2018w}
      • "Codd's theorem states that" the "following foundational query languages for the relational model are precisely equivalent in expressive power": relational algebra and domain-independent relational calculus queries \cite{WikipediaContributors2017a8}
    • object-oriented data model, object model, and objective database \cite{WikipediaContributors2018w}
    • entity-relationship model \cite{WikipediaContributors2018w}
      • enhanced entity-relationship model \cite{WikipediaContributors2018w}
    • document model \cite{WikipediaContributors2018w}
    • entity-attribute-value model \cite{WikipediaContributors2018w}
    • star schema \cite{WikipediaContributors2018w}
    • object-relational database \cite{WikipediaContributors2018w}
      • An object-relational database is a combination of the object model and the relational model.
  • rationale/justifications for creating logical data model \cite{WikipediaContributors2018y}:
    • Facilitates understanding of business data elements and requirements.
    • Provides the basis for database design.
    • Facilitates the avoidance of data redundancy, data inconsistency, and business transaction inconsistency; avoidance of the latter pair is dependent on avoiding data redundancy.
    • Facilitates re-use and sharing of data.
    • Facilitates the reduction of time and effort/cost in developing and maintaining the database.
    • Verifies the logical process models.
    • Facilitates impact analysis.
  • Includes entities (tables), attributes (columns/fields), and relationships (keys) \cite{WikipediaContributors2018y}
  • Uses defined (and specific) business names, and less generic names, for entities and attributes \cite{WikipediaContributors2018y}.
  • Is technology independent, and is based on platforms and DBMSes \cite{WikipediaContributors2018y}.
  • "Is normalized to the fourth normal form (4NF)" \cite{WikipediaContributors2018y}.
• physical data model, which describes how data is physically stored in terms of partitions, processors, and tablespaces.
  • Involves using specific database management technology \cite{WikipediaContributors2018y}.
  • "Includes tables, columns, keys, data types, validation rules, database triggers, stored procedures, domains, and access constraints" \cite{WikipediaContributors2018y}.
  • Uses more defined (and specific) business names, for entities and attributes, subject to limitations of DBMS and company defined standards \cite{WikipediaContributors2018y}; e.g., abbreviated column names \cite{WikipediaContributors2018y}.
  • "Includes primary keys and indices for fast data access" \cite{WikipediaContributors2018y}.
  • If the database is for "online transaction processing (OLTP) or Operational Data Store (ODS), it is usually not de-normalized"; else, de-normalize the physical data model to meet performance requirements of the database (context-dependent).
  • Examples:
    • inverted index \cite{WikipediaContributors2018w}
    • flat file \cite{WikipediaContributors2018w}

Other database models (or data models) \cite{WikipediaContributors2018w}:

• associative model
• correlational model
• multi-dimensional model
• multi-value model
• semantic model
• XML database
• named graph
• triplestore

WikipediaContributors2018z WikipediaContributors2018w

The relational data model is the most popular database model (and data model); it uses the table-based format \cite{WikipediaContributors2018w}.

Ancillary Definitions

A storage record is a group of fields, data, or words \cite{WikipediaContributors2016m}

• A record is "a self-contained collection of information about a single object", or a collection of distinct items \cite{WikipediaContributors2016m}.
• A row (or tuple) "represents a single, implicitly structured data item in a table," or "a set of related data" \cite{WikipediaContributors2018a15}
  • Each row in a table (or relational database) has the same structure \cite{WikipediaContributors2018a15}.
  • Each row in a table represents a set of related data \cite{WikipediaContributors2018a15}.
  • The implicit structure and significance of a row implies a sequence of data values (or a sequence of sets of data values) such that each column in that row has a value (or a set of data values) \cite{WikipediaContributors2018a15}.
  • A row is a relvar that is comprised of a set of two-tuples, where each two-tuple consist of the name of the appropriate column and the value associated with this row for that column \cite{WikipediaContributors2018a15}.
    • A relvar, or relation variable, is a variable that is assigned a relation and "the relation itself" \cite{WikipediaContributors2017a7}.
  • "A tuple is a finite sequence of attributes," where an attribute is an "ordered pair of domains and values" \cite{WikipediaContributors2018a25} [Here, tuple is defined with ordering.][Fix This!!!]
  • Each column of a row/table requires data value(s) of a specific data type \cite{WikipediaContributors2018a15}.
• A record (or structure, struct, or compound data) is a basic data structure \cite{WikipediaContributors2018a16}.
  • Rows of a database and spreadsheets are (also) known as records.
  • "Data structures serve as the basis for abstract data types (ADT). `The ADT defines the logical form of the data type. The data structure implements the physical form of the data type.' \cite{WikipediaContributors2018a17}"
    • "An abstract data type, a data structure that is defined indirectly by the operations that may be performed on it, and the mathematical properties of those operations (including their space and time cost)" \cite{WikipediaContributors2018a17}
    • "In computer science, an abstract data type (ADT) is a mathematical model for data types, where a data type is defined by its behavior (semantics) from the point of view of a user of the data, specifically in terms of possible values, possible operations on data of this type, and the behavior of these operations. This contrasts with data structures, which are concrete representations of data, and are the point of view of an implementer, not a user." \cite{WikipediaContributors2018a18}
    • "Formally, an ADT may be defined as a "class of objects whose logical behavior is defined by a set of values and a set of operations";[1] this is analogous to an algebraic structure in mathematics. What is meant by 'behavior' varies by author, with the two main types of formal specifications for behavior being axiomatic (algebraic) specification and an abstract model;[2] these correspond to axiomatic semantics and operational semantics of an abstract machine, respectively. Some authors also include the computational complexity ('cost'), both in terms of time (for computing operations) and space (for representing values)." \cite{WikipediaContributors2018a18}
  • A "record is a collection of fields" (or, members or elements), which are fixed in "number and sequence" and the fields can have different data types \cite{WikipediaContributors2018a16}.
    • Note that a record/element is a collection, not an element of a collection \cite{WikipediaContributors2018a16}.
    • A record is different from an array, since each field had a name and can be of a different type in comparison to other fields \cite{WikipediaContributors2018a16}
      • An array does not have names for each element, and all the elements of the array must have the same data type
      • This assumes that the array can be a dynamic array, which is also known as growable array, resizable array, dynamic table, mutable array, or array list.
  • A record type is a data type that describes a set of variables (i.e., the variable name is the identifier/label of the field) and their values \cite{WikipediaContributors2018a16}.
  • A record has a set of keys, which can be an empty set \cite{WikipediaContributors2018a16}.
    • A key is an identifier for a field, or a set of fields \cite{WikipediaContributors2018a16}.
• A record is a set of unordered pairs of a label and a value (or a set of label-accessible elements) \cite{WikipediaContributors2018a19}.
• A self-defining record is a self-contained collection of information that "identify the record type and locate information within the record", so that "elements can be stored in any order or omitted" \cite{WikipediaContributors2018a16}.

A relation is a set of tuples {d_1, d_2, ..., d_n}, where each tuple d_i is a member of a data domain D_i \cite{WikipediaContributors2018a1}.

• For any tuple of a relation, its elements (or attribute values) are not ordered \cite{WikipediaContributors2018a1}; note that, mathematically, a tuple is a finite ordered list/sequence of elements \cite{WikipediaContributors2018a19}.
  • An attribute (or data type or type) is a pair of a name and a domain \cite{WikipediaContributors2018a1}.
    • An attribute is an "ordered pair of domains and values" \cite{WikipediaContributors2018a25} [Are attributes supposed to be ordered?][Fix This!!!]
  • An attribute value is a pair of an attribute name and an element of the attribute's domain \cite{WikipediaContributors2018a1}.
  • A tuple is a set of unique attribute values, such that no pair of attribute values (or any two distinct attribute values) are the same \cite{WikipediaContributors2018a1}
    • A tuple can also be described as a function that maps names to values \cite{WikipediaContributors2018a1}.
    • "A tuple is a finite sequence of attributes" (i.e., "ordered pairs of domains and values") \cite{WikipediaContributors2018a25}
  • A heading is "a set of attributes" that has "no two distinct elements" [that] have the same name" \cite{WikipediaContributors2018a1}.
  • A body is a set of tuples that have the same heading \cite{WikipediaContributors2018a1}.
  • Hence, a relation is a pair of a heading and a body, such that "the heading of the relation" is "also the heading of each tuple in the body [of the relation]" \cite{WikipediaContributors2018a1}.
  • The degree of a heading is the number of attributes in the heading \cite{WikipediaContributors2018a1}.
  • The degree of a n-tuple is n, n >= 0 \cite{WikipediaContributors2018a1}.
• "A relation is a set of ordered pairs of domains and names" \cite{WikipediaContributors2018a25}; [a relation is a set of unordered tuples/(pairs of domains and values)][Fix This!!!]
  • Alternatively, "a relation is a set of (compatible) tuples" \cite{WikipediaContributors2018a25}.
• "In [the context of] database analysis and data management, a data domain refers to all the values [that] a data element [can] contain" \cite{WikipediaContributors2017a10}.
  • To determine the boundary of a data domain (or domain boundary), enumerate a list of values for a data type \cite{WikipediaContributors2017a10}.
  • This definition of data domain unites the following concepts of domain \cite{WikipediaContributors2017a10}:
    • An area that influence/control is exerted/asserted/exercised
    • "The set of values [for variables] for which [the] function/domain is defined"
• In terms of a finitary relation, under the closed-world assumption (CWA) \cite{WikipediaContributors2018a9}, a n-ary relation is a set of tuples on some set of n sets S_1, S_2, ..., S_n \cite{WikipediaContributors2018a1}.
  • This can be understood as an extension of a n-adic predicate, for all "n-tuples whose values, substituted for corresponding free variables in the predicate, yield propositions that hold true" implies and is implied by values in the relation \cite{WikipediaContributors2018a1}
  • "All and only spartans are bold" means the following.
    • $\forall x [B(x) \Leftrightarrow S(x)]$ \cite{rmw2015}
    • The logical biconditional "B(x) \Leftrightarrow S(x)" (if and only if) means that B(x) implies S(x) (B(x) \Rightarrow S(x)) and S(x) implies B(x) (B(x) \Leftarrow S(x), or S(x) \Rightarrow B(x))
  • "In set theory and [mathematical] logic, a relation is a mathematical property that assigns truth values to k-tuples of individuals" \cite{WikipediaContributors2018a20}.
    • "This property describes a possible connection between the components of a k-tuple" \cite{WikipediaContributors2018a20}.
    • "For a given set of k-tuples", assign "a truth value to each k-tuple" based on whether the property holds or not \cite{WikipediaContributors2018a20}.
  • A relation is an "ordered set" \cite{WikipediaContributors2018a20}.
  • The relation's arity, adicity, or dimension of a relation is k, and it is known as a k-ary relation, k-adic relation, or k-dimensional relation \cite{WikipediaContributors2018a20}.
  • "A n-ary (or k-ary) relation is a set of n-tuples" \cite{WikipediaContributors2018a20}.
  • If a relation has a finite arity, adicity, or dimension, it is known as a finite-place relation or finitary relation \cite{WikipediaContributors2018a20}.
  • We can generalize a finitary relation to an infinite sequence that includes infinitary relations between infinitudes of individuals \cite{WikipediaContributors2018a20}.
  • A relation over a collection of sets is a subset of their Cartesian product \cite{WikipediaContributors2018a20}.
  • When the sets are equivalent, the relation is homogeneous (i.e., homogeneous relation) \cite{WikipediaContributors2018a20}.
  • When each of the sets are unique, the relation is heterogeneous (i.e., heterogeneous relation) \cite{WikipediaContributors2018a20}.
    • For a relation over domains X_1, ..., X_k, a sequence of variables (x_1, ..., x_k) is a range over the respective domains (i.e., X_1, ..., X_k) \cite{WikipediaContributors2018a20}.
  • When the cardinality of the collection of sets is one, the property/relation is a unary relation \cite{WikipediaContributors2018a20}.
  • When the cardinality of the collection of sets is three, the property/relation is a ternary relation \cite{WikipediaContributors2018a20}.
  • When the cardinality of the collection of sets is four, the property/relation is a quaternary relation \cite{WikipediaContributors2018a20}.
  • Alternatively, an (embedded/included) relation is "a mathematical object determined by the specification of n component mathematical objects" (or n-tuples) \cite{WikipediaContributors2018a20}
    • For "a relation L over k sets, there are k+1 things to specify the k sets and a subset of their Cartesian product" -- "L is a (k+1)-tuple" \cite{WikipediaContributors2018a20}.
  • Each element set x_j of a relation is a domain of the relation \cite{WikipediaContributors2018a20}
    • The relation does not uniquely specify any given sequence of domains \cite{WikipediaContributors2018a20}
    • All domains x_js of a k-ary relation (over X) belong to the same set X \cite{WikipediaContributors2018a20}.
    • "If any domain x_j is empty, the only relation over such a sequence of domains is the empty relation L = empty set" since the defining Cartesian product is empty \cite{WikipediaContributors2018a20}.
    • For non-empty relations, none of the domains x_js can be empty \cite{WikipediaContributors2018a20}.
• A relation schema is a pair of a heading and "a set of constraints defined in terms of that heading", and can include a name \cite{WikipediaContributors2018a1}.
  • A relation is "an instantiation of a relation schema" if it has the heading of that [relation] schema and it satisfies the applicable constraints \cite{WikipediaContributors2018a1}.
  • A database schema (or "relational schema") is a collection of named relational schemas \cite{WikipediaContributors2018a1}.
  • Since "the domain of each attribute is a data type," and the named relational schema is effectively a relation variable (or relvar) \cite{WikipediaContributors2018a1}.
  • This is summarized from a set-theoretic perspective \cite{WikipediaContributors2018a20}.
  • "A relational database schema is a tuple S = (D, R, h), where D is the domain of atomic values, R is a finite set of relational names, h:R \rightarrow 2^{C} is a function that associates each header with its corresponding relation name in R, and C is a set of column names (or attributes) and includes headers as a finite subset of C \cite{WikipediaContributors2018a25}.
    • This definition uses a simpler relational model than the full relational model that has multiple domains and each header is a set of column names that is mapped to a domain \cite{WikipediaContributors2018a25}.
    • A tuple over a domain is a partial function that maps a set of column names tuples to an atomic value in D \cite{WikipediaContributors2018a25}.
    • Each tuple in a relation shall contain the same set of column names \cite{WikipediaContributors2018a25}.
  • A relational database is composed of named relation variables (or relvars), so that the database can be kept updated \cite{WikipediaContributors2018a1}.
    • When a relvar is updated, its body would be replaced by a different set of tuples \cite{WikipediaContributors2018a1}.
    • The two classes of relvar are: base relation variables and derived relation variables (or virtual relvars, referred to as the short-term view) \cite{WikipediaContributors2018a1}.
    • The "base relation variable is a relvar [that] is not derived from other relvars" \cite{WikipediaContributors2018a1}
      • The "base relvar" is independent from other relvars \cite{WikipediaContributors2018a1}.
    • The term base table in SQL is analogous to the base relation variable \cite{WikipediaContributors2018a1}.
    • A derived relation variable (virtual relvar, or view) is defined as a mathematical expression based on the operators of relational algebra or relational calculus \cite{WikipediaContributors2018a1}.
      • A derived relation variable operating on a set/collection of relations (assigned to database variables) results in another relation (i.e., "derived relation") \cite{WikipediaContributors2018a1}.
      • Each derived relation variable should contain at least one base relation variable as an operand \cite{WikipediaContributors2018a1}.
    • The Data Definition Language is used to define base relation variables and derived relation variables \cite{WikipediaContributors2018a1}.
• In SQL, a relation is a table representation, such that each row of the table represents a tuple and each column represents the values of an attribute \cite{WikipediaContributors2018a1}.
• The body of a relation has a set of unordered tuples \cite{WikipediaContributors2018a1}.
  • Similarly, the rows/records of an SQL table are unordered \cite{WikipediaContributors2018a1}.
• Similarly, the attributes/elements of a tuple/heading are unordered \cite{WikipediaContributors2018a1}.
• Additional notes
  • "In mathematics, a tuple is a finite ordered list (sequence) of elements" \cite{WikipediaContributors2018a19}.
  • "A n-tuple is an ordered list (or sequence) of n elements, where n is a non-negative integer" \cite{WikipediaContributors2018a19}.
  • A 0-tuple \cite{WikipediaContributors2018a20} is also known as an empty sequence, empty tuple \cite{WikipediaContributors2018a20}, null tuple, unit, or empty sequence \cite{WikipediaContributors2018a19}.
  • "There are only two zero-place relations": a 0-tuple that always holds, and the other 0-tuple that never holds \cite{WikipediaContributors2018a20}.
  • There exists only one instance of a 0-tuple \cite{WikipediaContributors2018a19}.
  • A 1-tuple is a singleton, monuple, or monad \cite{WikipediaContributors2018a19}.
  • A one-place relation is a unary relation, and a two-place relation is a binary relation (e.g., equalities, inequalities, divisors, or a set membership) or dyadic relation \cite{WikipediaContributors2018a20}.
    • Synonymous terms for a binary relation are: dyadic relation, 2-place relation, and correspondence \cite{WikipediaContributors2018a21}.
    • "A binary relation on A \times B is an element in the power set on A \times B" ("ordered by inclusion" in the lattice of subsets of A \times B) \cite{WikipediaContributors2018a21}.
    • A three-place relation (or 3-place relation) is also known as: ternary relation, triadic relation, 3-adic relation, 3-ary relation, or 3-dimensional relation \cite{WikipediaContributors2018a22}.
  • A 2-tuple is an ordered pair, dual, couple, twin, duad, or dyad \cite{WikipediaContributors2018a19}.
  • A 3-tuple is a triple, triplet, treble, or triad \cite{WikipediaContributors2018a19}.
  • A n-tuple can be defined as a function, nested ordered pair, nested sets \cite{WikipediaContributors2018a19}.

"A table is a collection of related data" stored in a database, using a structured format consists of (horizontal) rows and (vertical) columns (identifiable by name) \cite{WikipediaContributors2018a5,WikipediaContributors2018a15};

• the number of columns are specified (and fixed), and the table can have any number of rows \cite{WikipediaContributors2018a5};
• for any {row, column} entry, it can have multiple values \cite{WikipediaContributors2018a5};
• a table is defined for relational databases and flat-file databases \cite{WikipediaContributors2018a5}.
• A table can be used to describe a relation, which is a set without duplicates \cite{WikipediaContributors2018a5};
  • however, most tables are multi-sets (or bags) \cite{WikipediaContributors2018a5}.
    • E.g, this is true for SQL \cite{WikipediaContributors2018a26}.
  • For the relational model of database, "a table is [a] "visual representation of a relation" \cite{WikipediaContributors2018a25}
  • For the relational model of database, a table can be considered as a relation, even though they are not strictly equivalent \cite{WikipediaContributors2018a5,WikipediaContributors2018a25}.
• A table can have associated metadata, such as constraints on the table or values for certain columns \cite{WikipediaContributors2018a5}.

A view can function as a relational table, although its data would be computed/calculated at query time \cite{WikipediaContributors2018a5};

• an external table can be considered as a view \cite{WikipediaContributors2018a5}.

In a hierarchical database, which is a non-relational system, a table has a distant counterpart known as structured file, which can have repeating information in a row (i.e., the child data segments) \cite{WikipediaContributors2018a5};

• "data are stored [as a] sequence of physical records" \cite{WikipediaContributors2018a5}.

Notes about Codd's theorem:

• Either we can formulate a database query using relational algebra and domain-independent relational calculus, or it cannot be formulated/expressed \cite{WikipediaContributors2017a8}.
• Queries made using domain-independent relational calculus are invariant of selecting external domains (i.e., domains [of values] outside the database) \cite{WikipediaContributors2017a8}.
  • Database queries that "return different results for different domains" are forbidden, since they are domain dependent \cite{WikipediaContributors2017a8}.
  • We cannot perform database queries to "select all tuples" outside of relation R in the database" \cite{WikipediaContributors2017a8}.
  • To query a tuple constructed from "sets of atomic data items" is domain dependent (i.e., not domain independent), and yield different results \cite{WikipediaContributors2017a8}.
• Relational algebra and domain-independent relational calculus are fairly different foundational query languages, in terms of syntax, such that the former is a variable-free language and the latter is a logical language (related to first-order logic) with variables and quantification \cite{WikipediaContributors2017a8}.
• If a query language's expressive power is equivalent to that of relational algebra, it is relationally complete \cite{WikipediaContributors2017a8}.
  • E.g., relational calculus \cite{WikipediaContributors2017a8}.
  • Relational completeness does not guarantee all interesting database queries can be expressed in relationally complete languages \cite{WikipediaContributors2017a8}.
    • E.g., SQL operations for counting tuples and summing up values in tuples \cite{WikipediaContributors2017a8}.
    • E.g., Computing the transitive closure of a graph \cite{WikipediaContributors2017a8}.
    • This is because SQL has features that are not captured by relational algebra, such as \cite{WikipediaContributors2017a8}:
      • SQL nulls
      • three-valued logic
      • multi-set semantics
        • Can represent duplicate rows
  • Conflicting statement from \cite{WikipediaContributors2018a25} [FIX THIS!!!]
    • A relationally complete query language for the relational model can express queries in tuple relational calculus, domain relational calculus, and relational algebra.
    • Validate the aforementioned statement!!!
• Effectively, for expressing database queries in the relational model (for databases), relational algebra and relational calculus are logically equivalent \cite{WikipediaContributors2018a23}; for any expression in relational algebra, there exists an equivalent expression in relational calculus \cite{WikipediaContributors2018a23}.

Notes about relational calculus:

• Regarding the relational model of databases, relational calculus includes the following calculi to declaratively specify database queries \cite{WikipediaContributors2018a23}:
  • tuple relational calculus is a method of computation that serves as a declarative database query language for the relational model (or relational data model), regarding data manipulation \cite{WikipediaContributors2018a25}.
    • In tuple relational calculus, just like DRC, formulas use logical connectives (or logical operators) to combine atoms and use existential and universal quantifiers to bind variables \cite{WikipediaContributors2018a25}.
    • The aforementioned logical connectives are negation, conjunction, and disjunction \cite{WikipediaContributors2018a25}.
    • Tuple relational calculus contains fragments of first-order logic \cite{WikipediaContributors2018a25}.
    • For a given relational database schema, a query expression can be formed using formulas of the given schema \cite{WikipediaContributors2018a25}.
    • The results of domain- specific/dependent queries are proportional to the relational database schema's domain size \cite{WikipediaContributors2018a25}. Hence, use domain-independent queries instead of domain- specific/dependent queries, because domain-independent queries will return the same result for a given database, regardless of the used schema (and selected domain) \cite{WikipediaContributors2018a25}.
  • domain relational calculus (DRC)
    • Domain relational calculus (DRC) is a method of computation that serves as a declarative database query language for the relational model (or relational data model) \cite{WikipediaContributors2017a9}.
    • Queries in DRC have the structure/form of a set of domain variables or constants, such that the set DRC formulas in that structure are true \cite{WikipediaContributors2017a9}.
    • DRC and tuple relational calculus have the same operators: logical connectives, such as negation, conjunction, and disjunction \cite{WikipediaContributors2017a9}.
    • In DRC, variables can be bind with existential and universal quantifiers \cite{WikipediaContributors2017a9}.
    • domain relational calculus (DRC) is also known as domain calculus \cite{WikipediaContributors2018a25}
    • DRC contains fragments of first-order logic \cite{WikipediaContributors2018a25}.
• In terms of expressive power, tuple relational calculus and domain relational calculus are equivalent \cite{WikipediaContributors2018a25}.
• The computational expressiveness of DRC and relational algebra are equivalent \cite{WikipediaContributors2017a9}.
• In terms of expressive power, tuple relational calculus and relational algebra are equivalent \cite{WikipediaContributors2018a25}.
• In contrast, the relational model of databases allows relational algebra to provide procedural specification of database queries \cite{WikipediaContributors2018a23}.
• Note that the definition of calculus is "any method or system of calculation" \cite{WikipediaContributors2018a24}.

Notes about relational algebra:

• The relational model of databases allows relational algebra to provide procedural specification of database queries \cite{WikipediaContributors2018a23}.
• In contrast, relational model of databases allows relational calculus to declaratively specify database queries \cite{WikipediaContributors2018a23}.
• "Relational algebra is a family of algebras" that has well-founded semantics for operations on relational databases, such as modeling data storage "and defining queries" \cite{WikipediaContributors2018a26}.
  • The primary/primitive operators of Codd's algebra (from E. F. Codd's relational model) form the basis of query languages for database management \cite{WikipediaContributors2018a26}.
    • The operators are \cite{WikipediaContributors2018a26}:
      • selection
      • projection
      • Cartesian product (or, cross product, or cross join)
      • set union
      • set difference
  • The natural join SQL operator, and join-like SQL operators, enables compositions of relations to be defined \cite{WikipediaContributors2018a26}.
    • In category theory \cite{WikipediaContributors2018a29,WikipediaContributors2018a30}, the join operator is known as fiber product \cite{WikipediaContributors2018a26}, pullback, fibre product, fibered product, or Cartesian square \cite{WikipediaContributors2018a27}.
      • In category theory \cite{WikipediaContributors2018a30}, limits \cite{WikipediaContributors2018a28} (such as pullbacks) represent the inherent/intrinsic/essential properties of universal constructions \cite{WikipediaContributors2018a27,WikipediaContributors2018a29}.
      • Category theory \cite{WikipediaContributors2018a30} is a part of abstract algebra \cite{WikipediaContributors2018a31}.
    • The natural join SQL operator enables relations associated with a foreign key \cite{WikipediaContributors2018a34} to be combined \cite{WikipediaContributors2018a26}.
  • Relational algebra cannot express transitive closures \cite{WikipediaContributors2018a26}; relational algebra does not allow expressions to represent operators, such as transitive closure \cite{WikipediaContributors2018a26}.
    • However, SQL allows fixpoint queries \cite{WikipediaContributors2018a26} and general recursive fixpoint queries, such as hierarchical queries \cite{WikipediaContributors2018a32}.
      • hierarchical queries are implemented by recursive common table expressions (CTE) \cite{WikipediaContributors2018a32}
  • Use properties of relational algebra to perform query optimization \cite{WikipediaContributors2018a26}
    • A binary expression tree is a special type of binary trees that represent algebraic and boolean expressions \cite{WikipediaContributors2018a33}.
    • Represent queries as binary expression trees, such that their "internal nodes are operators", "leaves are relations", and "subtress are subexpressions" \cite{WikipediaContributors2018a26}.
    • Query optimization transforms a binary expression trees into equivalent but smaller binary expression trees \cite{WikipediaContributors2018a26}, such that their subtrees are smaller.
      • A smaller subtree implies smaller subexpressions that provide/produce smaller relations \cite{WikipediaContributors2018a26}.
      • In addition, query optimization tries to compute common subexpressions once, and reuse these results in other queries that contain these common subexpressions \cite{WikipediaContributors2018a26}.
      • Rules for query optimization \cite{WikipediaContributors2018a26}:
        • selection
        • projection
        • rename

"A database segment is a database object that occupies physical space, such as table data and indexes/indices" \cite{WikipediaContributors2018z}.

A tablespace is a storage location of the actual data underlying database objects (database storage locations) \cite{WikipediaContributors2018z}.

• it provides a layer of abstraction between the logical and physical data models \cite{WikipediaContributors2018z};
• it allocates storage for all data segments managed by the DBMS \cite{WikipediaContributors2018z};
• when creating database segments, we can refer to the tablespace by name \cite{WikipediaContributors2018z};
• it does not store the logical database structure;
  • for a given logic schema, an unique object in the schema has a unique tablespace \cite{WikipediaContributors2018z};
  • for a given tablespace, it allows multiple database segments to refer to it \cite{WikipediaContributors2018z};
  • for a given tablespace, use it to specify a database model that forms a bond between logical and physical data \cite{WikipediaContributors2018z};
  • use a tablespace to optimize performance of database access/modification and decide where to store indexes/indices and tables \cite{WikipediaContributors2018z};
  • a tablespace can store its data in a file in the file system \cite{WikipediaContributors2018z};
  • a file cannot be associated with multiple tablespaces \cite{WikipediaContributors2018z};
  • a DBMS allows the direct configuration of a tablespace over device entries of an operating system (i.e., raw devices), in order to gain a performance speedup "by avoiding OS file system overheads" \cite{WikipediaContributors2018z}.

In UNIX-like operating systems, a raw device is a special logical device that is associated with character device files \cite{WikipediaContributors2018a3,Bovet2006}, and enables/allows direct access by a storage device (e.g., hard disk drive) \cite{WikipediaContributors2017a6}.

• that is, the raw device allows software applications to use storage devices directly, without using the page caches \cite{WikipediaContributors2018a4} and buffers of the operating system - although the disk buffer \cite{WikipediaContributors2018a2,WikipediaContributors2016l} of the tertiary storage devices would still be used \cite{WikipediaContributors2017a6}.

In UNIX-like operating systems, a device file (or special file) is an interface to a device driver, and appears in a file system as an ordinary file \cite{WikipediaContributors2018a3}.

• Using I/O system calls for the application, users can interact with its device driver \cite{WikipediaContributors2018a3}.
• It is managed by the virtual file system \cite{WikipediaContributors2018a3};
  • the controlling daemon "monitors hardware addition and removal at run time" and modifies the device file system (if the device file system has not been modified by the kernel) \cite{WikipediaContributors2018a3}.

A "character device (driver), or character special file, provides unbuffered, direct access to the hardware device" \cite{WikipediaContributors2018a3};

• it can also request for read and write operations to align to block boundaries (or otherwise) \cite{WikipediaContributors2018a3};
• block-based hardware typically requires software to read/write aligned blocks \cite{WikipediaContributors2018a3}.

A block device, or block special file, provides software with buffered access to hardware devices with restrictions on size or alignment \cite{WikipediaContributors2018a3};

• however, it has no guarantee on performance nor order of data between any character, byte, nor block, due to the buffering \cite{WikipediaContributors2018a3}.

An operating system can represent hardware devices, such as hard disks, as character/block devices \cite{WikipediaContributors2018a3}.

A device node corresponds to a resource allocated by the kernel of the operating system \cite{WikipediaContributors2018a3}.

A pseudo-device is a device node in UNIX-like systems that do not correspond to a physical device \cite{WikipediaContributors2018a3}.

The mknod system call, which is a service request made from a computer program on the OS kernel, creates nodes in the file system tree \cite{WikipediaContributors2018a3}; such nodes can be moved or deleted using file system system calls and commands \cite{WikipediaContributors2018a3}.

A query language, or data query languages (DQL), is a computer language for making queries on databases and information systems \cite{WikipediaContributors2018v}.

The main categories of query languages are \cite{WikipediaContributors2018v}:

• database query languages
• information retrieval query languages

A primary key is a specific choice of columns that can uniquely identify rows \cite{WikipediaContributors2018a5} (or identify a tuple in a relation \cite{WikipediaContributors2018a6}).

• For "the relational model in databases, a primary key is a specific choice of a minimal set of attributes (columns) that uniquely specify a tuple (row) in a relation (table)" \cite{WikipediaContributors2018a6};
• Mathematically, "a primary key is a choice of candidate key (i.e., a minimal superkey), and other candidate keys are alternate keys \cite{WikipediaContributors2018a6}.
  • An alternate key is also known as a secondary key \cite{WikipediaContributors2018a16}.
• Assign an unique index to each alternate key, so that we can use the unique index to determine if duplicates (of alternate keys) exist \cite{WikipediaContributors2018a6};
  • Do/Use this to prevent insertion/addition of duplicate alternate keys \cite{WikipediaContributors2018a6}; and
  • Unique columns in databases cannot include duplicates \cite{WikipediaContributors2018a6}.
  • For a single-table select, or filtering in a where clause, we can use alternate keys as primary keys \cite{WikipediaContributors2018a6}.
  • However, alternate keys cannot be used as primary keys when joining multiple tables \cite{WikipediaContributors2018a6}.
• A primary key is a unique key that is also known as a record key \cite{WikipediaContributors2018a16}.

A natural key is a primary key consisting of real-world observables \cite{WikipediaContributors2018a6}.

• A natural key (or, business key or domain key) is a unique key in database design that uses the relational model, and is based on real-world attributes \cite{WikipediaContributors2018a7}.
• Use natural keys in business-related columns \cite{WikipediaContributors2018a7}.
• "A natural key is a candidate key that has a logical relationship to the attributes within that row" \cite{WikipediaContributors2018a7}.
• The primary advantage of a natural key over a surrogate key is its existence \cite{WikipediaContributors2018a7};
  • hence, no new, artificial column has to be added to the schema to create the surrogate key \cite{WikipediaContributors2018a7}.
  • this has no significant meaning outside the database environment \cite{WikipediaContributors2018a7}.
  • When a natural key can be identified, selection of the natural key over the surrogate primary keys can simplify data processing \cite{WikipediaContributors2018a7}.
  • It ensures that there exists only a row per key, since this is based on a real-world observation \cite{WikipediaContributors2018a7}.
• I noted some flimsy connection to the following references:
  • Open-world assumption (OWA) \cite{WikipediaContributors2018a8}
  • Closed-world assumption (CWA) \cite{WikipediaContributors2018a9}
  • Single version of the truth (SVOT) \cite{WikipediaContributors2018a10}
    • A data warehouse that has a single centralized database, or a distributed synchronized database, to store "all of an organization's data in a consistent and non-redundant form."
  • Single source of truth (SSOT) \cite{WikipediaContributors2018a11}
    • A method to "structure information models and associated data schema," so "that every data element is stored exactly once" \cite{WikipediaContributors2018a11}.
    • "Always source a particular piece of information from one place" \cite{WikipediaContributors2018a10}.
    • Use referencing to link to data elements in the SSOT databases, which are based on the relational schema or distant federated databases, so that "all other locations of the data" refer/point "to the primary `source of truth' location" \cite{WikipediaContributors2018a11};
      • this avoids the need to keep duplicates of a data element updated, by allowing value from the only/primary location to propagate throughout the computer network \cite{WikipediaContributors2018a11};
      • it also avoids the risks of "incorrectly linked duplicates", and denormalizing data elements (see database denormalization \cite{WikipediaContributors2018a12}).
      • Usage of pointers also covers copying and updating database tables, rows, and cells \cite{WikipediaContributors2018a11}

A surrogate key is an attribute that functions as a key \cite{WikipediaContributors2018a6,WikipediaContributors2018a13}.

• Synonyms of "surrogate key" are \cite{WikipediaContributors2018a13}:
  • synthetic key
  • entity identifier
  • system-generated key
  • database sequence number
  • factless key
  • technical key
  • arbitrary unique identifier
• In certain circumstances, during software development or a data science project, it can be inconvenient to use a natural key to identify a tuple in a relation, since it may require multiple columns or large text fields \cite{WikipediaContributors2018a6}.
  • Hence, use a surrogate key as a substitute (primary key) to avoid giving more priority/importance to the natural key or other candidate keys \cite{WikipediaContributors2018a6}.
  • Similarly, a surrogate key can be chosen instead of any of the multiple candidate keys that do not provide an advantage over the other candidate keys \cite{WikipediaContributors2018a6}.
• Since primary keys are chosen to facilitate information processing during software development or a data science project, many cases of database application design use surrogate primary keys to further facilitate information processing \cite{WikipediaContributors2018a6}.
• For databases based on the hybrid object-relational model (ORM), which is based on the object-oriented programming model and the relational model, they also use surrogate primary keys to further facilitate information processing \cite{WikipediaContributors2018a6}.
  • The restrictions on surrogate primary keys are \cite{WikipediaContributors2018a6}:
    • Primary keys are immutable (not changed nor reused) \cite{WikipediaContributors2018a6}
    • Primary keys should be deleted, together with associated record \cite{WikipediaContributors2018a6}
    • Primary keys should be an anonymous trigger, or numeric identifier \cite{WikipediaContributors2018a6}
    • These restrictions only apply for the object-relational model, such as the active record pattern \cite{WikipediaContributors2018a6}
    • Hence, for databases based on the relational model, or SQL standard, do due diligence when deciding which key should be an immutable primary key \cite{WikipediaContributors2018a6}; some DBMSes do not allow usage of the UPDATE SQL statement to change values of the primary keys \cite{WikipediaContributors2018a6}.
• A surrogate key "is a unique identifier for an entity in the modeled world ([data model]) or an object in the database ([storage model])" \cite{WikipediaContributors2018a13}.
  • It "is not derived from application data" \cite{WikipediaContributors2018a13}
    • In contrast, a natural key is derived from application data \cite{WikipediaContributors2018a13}.
  • A storage model represents important physical aspects of a data store's data structure \cite{WikipediaContributors2011};
    • on the other hand, a data model represents important logical aspects of a database's data structure \cite{WikipediaContributors2011}
• The attributes/features/qualities/characteristics of a surrogate key are \cite{WikipediaContributors2018a13}:
  • a unique, system-wide value that is never reused
  • system generated value; it is generated by the database management system
  • immutable value that cannot be modified by users or applications
  • value without semantic significance
  • value is not accessible by users or applications
  • value is not comprised/combined/composed of values from different domains, and cannot be decomposed into constituents, and not a synthesis/derivation of application data in the database.
  • a surrogate key can be used as a primary key
• The surrogate key can exist as a separate from other database/system generated values, such as universally unique identifier (UUID) and globally unique identifier (GUID) \cite{WikipediaContributors2018a13,WikipediaContributors2018a14}.
• Surrogate keys are typically sequential numbers \cite{WikipediaContributors2018a13}.
• By designing the surrogate key to be independent from all fields of the database, changes in the data values or design of the database would not affect the value of the surrogate key \cite{WikipediaContributors2018a13}; such designs of the database facilitate software development using agile development processes and ensure that the surrogate keys remain unique \cite{WikipediaContributors2018a13}.
• Advantages of surrogate keys \cite{WikipediaContributors2018a13}:
  • immutability of surrogate keys, unlike primary keys and natural keys
  • requirement changes, which may affect natural keys, would not affect surrogate keys; merging databases may affect natural keys, but not surrogate keys
  • performance (in terms of lookup time... access time [FIX THIS!!!]) of surrogate keys is better than natural/business keys
    • This is because the former only depends on finding records with one column (unique, immutable surrogate key), while the latter depends on finding records with multiple columns \cite{WikipediaContributors2018a5,WikipediaContributors2018a6}.

In the context of relational databases, a foreign key is a non-empty collection of fields (including a single field) in a table (Table A) that uniquely identifies a row of another table (Table B, or the same table Table A) \cite{WikipediaContributors2018a34}.

• That is, a foreign key, which is defined in Table B, refers to a unique key in Table A \cite{WikipediaContributors2018a34}.
• The table defining the foreign key is the child table, and the other table using the foreign key to uniquely identify a row is known as the referenced table (or parent table).
  • The following terms are used interchangeably \cite{WikipediaContributors2018a34}:
    • referenced table
    • parent table
    • master table
  • The following terms are used interchangeably \cite{WikipediaContributors2018a34}:
    • child table
    • referencing table
  • A one-to-many relationship between the parent table (or referenced table) and the child table (or referencing table) can exist \cite{WikipediaContributors2018a34}.
  • When the child table (referencing table) and the parent table (or referenced table) are the same, the foreign key refers back to the same table \cite{WikipediaContributors2018a34}.
    • Foreign keys that refer back to the same table is known in SQL:2003 (current revision of the SQL database query language is SQL:2016) are also known as the following \cite{WikipediaContributors2018a34}:
      • self-referencing foreign key
      • recursive foreign key
    • "In database management systems, this is often accomplished by linking a first and second reference to the same table"??? [Fix This!!!] \cite{WikipediaContributors2018a34}
• The referential integrity constraint for the child table and the referenced table (or parent table) is that the foreign key of the referenced table (or parent table) must be equivalent to the candidate key for a particular row of the child table (or primary table??? [Fix This!!!]) \cite{WikipediaContributors2018a34}.
  • The foreign key of the referenced table (or parent table) cannot be a candidate key for multiple rows of the child table/relation \cite{WikipediaContributors2018a34}.
  • Many "database management systems provide mechanisms" to check/validate that the referential integrity constraint is satisfied \cite{WikipediaContributors2018a34}.
  • When referential integrity constraint is not satisfied, the foreign keys liable for violating this constraint in the referenced table (or parent table) can be invalidated or modified \cite{WikipediaContributors2018a34}.
• When designing databases, foreign keys facilitate the modeling of relationships between different entities \cite{WikipediaContributors2018a34}.
• After carrying out database normalization \cite{WikipediaContributors2018a35} to remove data redundancy and to improve data integrity, foreign keys can connect (separate) tables in the database that were separated during database normalization \cite{WikipediaContributors2018a34}.
• A child table (or referencing table) can have multiple foreign keys, and each foreign key can have a different parent table (or referenced table) \cite{WikipediaContributors2018a34}.
  • The database management system will enforce each foreign key independently \cite{WikipediaContributors2018a34}.
  • Hence, foreign keys allow "cascading relationships between tables" to be formed \cite{WikipediaContributors2018a34}.
• Use data integrity constraints to define foreign keys in the \cite{WikipediaContributors2018a34}.
  • Use the ALTER TABLE statement to exclude/omit references in the child table (or referencing table), so that the foreign keys can reference the primary keys in the parent table (or referenced table) \cite{WikipediaContributors2018a34}.
  • Alternatively, define foreign keys via the CREATE TABLE statement, and include references in the child table (or referencing table) to link to (or reference) the primary keys in the parent table (or referenced table) \cite{WikipediaContributors2018a34}.
  • When rows/records in a parent table (or referenced table) are deleted or updated, carry out (one of) the following five referential actions to foreign keys in the child table (or referencing table) \cite{WikipediaContributors2018a34}:
    • CASCADE
    • RESTRICT
    • NO ACTION
    • SET NULL
    • SET DEFAULT

Other types of keys:

• "A composite key is candidate key that is comprised of [multiple] attributes (table columns) that uniquely identify an entity occurrence (table row)" \cite{WikipediaContributors2018a36}.
• A compound key is a composite key, where each attribute is a foreign key \cite{WikipediaContributors2018a36}.
  • A compound key is a composite key that consists of foreign keys as attributes \cite{WikipediaContributors2018a36}.

In a temporal database, each row has a natural/business key and the surrogate key, so that the former has a mapping to an unique entity in modeled world and the latter has a mapping to a unique row in the database \cite{WikipediaContributors2018a13}.

• Each row in the table/database represents the values of attributes for a time slice, and indicate the life span of the entity \cite{WikipediaContributors2018a13}.
• Hence, for a natural/business key or unique entity in modeled world, it can have multiple entries in a temporal database (since a person or an organization can have multiple business entities) \cite{WikipediaContributors2018a13}.

"In relational database terms, a primary key does not differ in form or function from alternate keys" \cite{WikipediaContributors2018a6};

• there can be different reasons for choosing a key as the primary key over other keys, such as \cite{WikipediaContributors2018a6}:
  • it is the "preferred" identifier for data in the table;
  • its usage as foreign key references \cite{WikipediaContributors2018a34} from other tables;
  • it indicates certain technical rather than semantic feature of the table; and
  • special syntax features of certain computer languages and software that are used to identify primary keys.

Even though the relational database model (based on relational calculus and relational algebra) does not distinguish keys based on whether they are primary keys, the SQL computer language standard has a feature for primary keys to provide a convenience to the application engineer \cite{WikipediaContributors2018a6}.

Side Notes on Definitions

A domain model captures concepts of a problem domain, but it does not capture the relationships (and their structure) of data in that domain; a logical data model does capture such relationships and their structure \cite{WikipediaContributors2018y}.

Database Design

cite this!!!

• Database design is the process of organizing data/information, using a database model.
• Database design involves creating the ontology of the desired data set:
  • Data classification
    • Deciding what/which data to store.
  • Identifying the relationships between the data.
• The types of ontology are:
  • domain ontology (or domain-specific ontology)
  • upper ontology (or foundation ontology)
    • A model of common objects, and common relationships between these objects, that can be applies to a set of domain ontologies.
  • hybrid ontology
    • A combination of domain ontology and upper ontology.
  • Reference: https://en.wikipedia.org/wiki/Ontology_(information_science)
    • ontology visualization techniques
    • ontology engineering (or ontology building), part of knowledge engineering

object-relational mapping

Database Models

Categories of Databases

Object Databases

SQL

Relational databases

• Components of relational databases, and their analogous mapping to \cite{WikipediaContributors2018a25}:
  • The basic building block of a relational database is the domain (or data domain), which is "similar [to], but not equal to, a data type \cite{WikipediaContributors2018a25}.
    • See Ancillary Definitions regarding the definitions of domain (or data domain).
  • row, which is analogous to a tuple \cite{WikipediaContributors2018a25}
    • See Ancillary Definitions regarding the definitions of row and tuple.
  • table, which is a "visual representation of a relation" \cite{WikipediaContributors2018a25}
    • See Ancillary Definitions regarding the definitions of table and relation.
• Quasi-relational SQL
  • Even though relational algebra underpins SQL, SQL does not implement accurately/exactly \cite{WikipediaContributors2018a26}.
    • For example, tables (or operands) in SQL do not correspond to (or concur with) relations in relational algebra \cite{WikipediaContributors2018a26}.
      • A SQL table is a multi-set (or bag), as opposed to a set (of tuples, or ordered pairs of domains and names) in relational algebra \cite{WikipediaContributors2018a26}.
    • SQL implementations do not allow some theorems in relational algebra to be specified \cite{WikipediaContributors2018a26}.
• relational database management systems (RDBMS)
  • database servers
    • MySQL
    • PostgreSQL
      • Supports ODBC
    • SQLite
    • Firebird (database server)
  • database clients
    • SQuirrel SQL client
  • distributed SQL
    • Apache Ignite: https://en.wikipedia.org/wiki/Apache_Ignite

NoSQL

Key-Value Store

NoSQL Database Management Systems

• HBase
• Cassandra
• MongoDB

Document-oriented database (= document store)

• Use document-oriented information (= semi-structured data)
• Examples:
  • XML database

NewSQL

Distributed Databases for Big Data

• Hive
• Spark
• Kafka
• Flume

Apache Ignite [FIX THIS!!!]

Distributed Databases Based on MapReduce

Apache Hadoop (and HDFS):

• Pig

Database Data Formats

Data formats for databases \cite{WikipediaContributors2018r}:

• SQL
• ODBC, Open Database Connectivity \cite{WikipediaContributors2018t}
  • API to access DBMS
  • There exists ODBC-to-JDBC (ODBC-JDBC) and JDBC-to-ODBC (JDBC-to-ODBC) bridges.
  • Also, see unixODBC and Microsoft Windows ODBC \cite{WikipediaContributors2018u} for the ODBC data API for associated operating systems.
• JDBC, Java Database Connectivity.
  • API based on Java to access/modify a database \cite{WikipediaContributors2018s}.
• GDA, GNU Data Access \cite{WikipediaContributors2017a5}
  • GNOME-DB is a GNOME-based database management systems.
  • It supports access to persistent data (in databases).
  • GDA, GNU Data Access, is its data management API.
    • Compared to JDBC and ODBC, it provides a larger set of features, and is considered as a complete architecture for databases.
    • "Libgda is a database access library", which serves as "a database and abstraction layer".

Use ODBC, JDBC, and GDA wrappers for database management systems of my choice \cite{WikipediaContributors2017a5}.

To-Do List: Tasks

• Use design of experiments (DOE) to access quality of database systems
• Enumerate list of database environments mentioned in the 2020 Stack Overflow Developer Survey
  • Cassandra
  • Couchbase
  • DynamoDB
  • Elasticsearch
  • Firebase
  • IBM DB2
  • MariaDB
  • Microsoft SQL Server
  • MongoDB
  • MySQL
  • Oracle database platform
  • Redis
  • SQLite

References

Citations/References that use the LaTeX/BibTeX notation are taken from my BibTeX database (set of BibTeX entries).

If these citations/references are not found in this list of references, information about them can be found in my BibTeX database.

Author Information

The MIT License (MIT)

Copyright (c) <2017> Zhiyang Ong

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Email address: echo "cukj -wb- 23wU4X5M589 TROJANS cqkH wiuz2y 0f Mw Stanford" | awk '{ sub("23wU4X5M589","F.d_c_b. ") sub("Stanford","d0mA1n"); print $5, $2, $8; for (i=1; i<=1; i++) print "6\b"; print $9, $7, $6 }' | sed y/kqcbuHwM62z/gnotrzadqmC/ | tr 'q' ' ' | tr -d [:cntrl:] | tr -d 'ir' | tr y "\n" Don't compromise my computing accounts. You have been warned.