Databases

  • Databases are structured sets of data that are stored electronically in a computer system.
  • The Database Management System (DBMS) is responsible for managing the interface between the data and the end user. Some popular DBMS examples are SQL, Oracle, and MySQL.
  • A database can either be a single table (flat-file database), or multiple tables linked together (relational database).
  • Databases handle data in a way that ensures ACID properties - Atomicity, Consistency, Isolation, and Durability.
  • A relational database uses tables (relations) to store data. Each row (tuple) represents a single record, while columns (attributes) store the record’s data elements.
  • SQL (Structured Query Language) is a standard language used to manage and manipulate relational databases.
  • Primary keys are unique identifiers for each record in a database. Foreign keys link records in two tables together, enabling relational functionality.
  • Normalisation is the process of reducing data redundancy within a database to improve its efficiency.
  • Indexing is a DBMS feature that improves the speed of data retrieval operations on a database by providing rapid random and sequential access to rows in a table.
  • In databases, a transaction refers to a single logical unit of work that is either entirely completed or not at all. If a transaction is interrupted, all changes are rolled back.
  • Databases allow concurrent transactions to take place to boost efficiency. However, this results in conflicts, which is where ACID properties are crucial.
  • Database security is vital to protect data from unauthorised access and data corruption during its lifecycle.
  • Backup and recovery processes are crucial for dealing with system crashes and data losses. Backups can be full, differential, or incremental.
  • Mirroring and replication are two techniques used for improving database availability, performance, and data protection with redundancy.
  • Big data describes large volumes of data that traditional database systems can’t handle. Three V’s that characterise big data are Volume, Variety, and Velocity.