Database management system (DBMS)

About

A DBMS ¹⁾ is an application that controls:

• the organization,
• storage,
• management,
• and retrieval of data.

A database is a collection of permanently stored data used by an application or enterprise. A database contains logically related data, which means that the database was created to represent the real-world.

All databases are not built on the relational model, see Database - Model

A collection of information organized to afford efficient retrieval

They are also known as storage engine.

Example

• A file system is a DBMS for a file

API

• ODBC - Open Database Connectivity
• JDBC (Java Database Connectivity)
• Database - Call level interface (CLI)

Properties

• Data Property - ACID (atomicity, consistency, isolation, durability)
• Distributed Database - Eventual consistency (Weak)
• Data Property - BASE (Basically Available)
• Distributed Database - CAP Theorem (Consistency, Availability, Partition Tolerance)
• Concurrency - Lock (Mutex)
• Data Processing - Replication
• Concurrency - Concurrency
• SQL Engine - Query Optimizer (Query Optimization)
• Data - Transaction (Trans(versal?) actions)

Mutable vs Immutable

After writing data:

• Mutable: the data can be overwritten later in the future (ie state-driven application (business process system, state machine)
• Immutable: the data can only be read again (ie event-driven data application)

Key Features

Why would I want a database? What problem do they solve?

1. Sharing. Support concurrent access by multiple readers and writers
2. Data Model Enforcement. Make sure all applications see clean, organized data
3. Scale. Some work with datasets too large to fit in memory
4. Flexibility. Use the data in new, unanticipated ways. Abstraction … Layer between physical and logical.

Documentation / Reference

• Bill Howe, PhD (Director of Research, Scalable Data Analytics) University of Washington - eScience Institute