OLTP handles operational transactions; OLAP supports analysis. ETL moves and transforms data, while replication maintains copies. Similar-looking database terms become easier to read when you ask whether the work is a transaction, an analytical query, a data pipeline, or an operational safeguard.
Core words
The diagram separates transaction processing, analytical stores, data movement, and database operations.
| English term | Meaning and use |
|---|---|
Data warehouse | An integrated store of data from multiple sources organized to support reporting and analysis. |
Big data | Data whose volume, variety, or velocity challenges conventional storage and processing approaches in a given context. |
Data mining | Analyzing large data sets to discover patterns, relationships, or useful predictive signals. |
DSMS | Data Stream Management System: software for continuously processing data arriving as streams. |
DC | Dublin Core: a standard set of metadata elements for describing resources; here DC does not mean data center. |
MDR | Metadata Registry: a managed repository of metadata definitions and their relationships. |
Database Tuning | Measuring and adjusting queries, indexes, schema, configuration, or resources to improve database performance. |
EA | Enterprise Architecture: a structured description of an organization's business, information, application, and technology architecture. |
Words to distinguish together
| English term | Meaning and use |
|---|---|
ERP | Enterprise Resource Planning: integrated software supporting organizational resources and functions such as finance, production, and logistics. |
DRM | Digital Rights Management: controls intended to manage or restrict access and use of digital content. |
OLAP | Online Analytical Processing: query and analysis of data for exploration, aggregation, and decision support. |
OLTP | Online Transaction Processing: handling many operational transactions such as orders, payments, or account updates. |
Data Mart | A focused subset of analytical data for a department, subject, or user group. |
Ontology | A formal model of concepts, properties, and relationships within a knowledge domain. |
meta data | Metadata: data describing other data, such as a column definition, source, or format. |
ETL | Extract, Transform, Load: a pipeline pattern for obtaining source data, changing it, and loading it into a target system. |
Words encountered in code and operations
| English term | Meaning and use |
|---|---|
LOB | Large Object: a database type or value for substantial binary or text content. |
work load | Workload: the volume and characteristics of tasks a system processes over time. |
Database Replication | Maintaining copies of database data, often for availability, read capacity, or recovery, with consistency and lag to consider. |
Clustering | Combining nodes into a service arrangement for availability or scale; shared storage is only one possible implementation. |
Precompile | A preparation step before compilation, such as source transformation or processing embedded SQL; exact use depends on the tool. |
Cursor | A database interface for traversing or processing rows from a query result. |
MyBatis | An open-source SQL mapping framework for Java that simplifies database access while retaining explicit SQL. |
Reading these terms in context
A data warehouse integrates data for analysis, often from operational sources. A data mart serves a narrower audience or subject. Replication can help availability or read scaling, but lag and consistency still need design. Clustering is a broader deployment arrangement and is not simply another word for replication.
ETL loads curated data for OLAP; OLTP continues to process day-to-day transactions, and replication maintains additional copies.
Key takeaways
Sort terms by purpose before choosing a tool: OLTP for operational transactions, OLAP for analysis, ETL for movement and transformation, metadata for describing data, and tuning for measured performance work. Neither a replica nor a cluster automatically fixes a slow query.
