AWS Database key terms

A MySQL- and PostgreSQL-compatible relational engine on AWS that stores data in a distributed, self-healing cluster volume with six copies spread across three Availability Zones.

An Aurora configuration that scales compute capacity up and down in fine-grained Aurora Capacity Units based on current load, without provisioning or managing fixed instance sizes.

A read-only copy of a database that receives asynchronous updates from the primary instance and is used to offload read traffic and improve read scalability.

A configuration in which the database keeps a synchronous standby copy in a second Availability Zone and fails over to it automatically when the primary becomes unavailable.

The attribute whose hashed value decides which physical partition stores an item; it must be chosen so that traffic and data spread evenly across partitions.

An index with its own partition and sort key that lets you query a table on alternate attributes; it is maintained asynchronously and has separately provisioned throughput.

DynamoDB throughput units: one read capacity unit covers one strongly consistent read per second for an item up to 4 KB, and one write capacity unit covers one 1 KB write per second.

An ordered, time-limited log of item-level changes in a DynamoDB table that applications or Lambda functions can consume to react to modifications.

A table setting that automatically deletes expired items based on a timestamp attribute, freeing storage without consuming write capacity.

A managed, columnar, massively parallel data warehouse service built for analytical SQL queries over large datasets.

A Redshift capability that runs SQL queries directly against files stored in Amazon S3, extending warehouse analysis to external data without loading it first.

A managed in-memory caching service supporting Redis OSS or Memcached engines, used to cut database load and reduce latency for repeated reads.

A managed document database service compatible with MongoDB APIs that stores JSON-like documents and supports indexing and rich query operations.

A managed graph database service optimized for storing and querying highly connected data using Gremlin, SPARQL, or openCypher.

A managed, serverless, wide-column database service compatible with Apache Cassandra Query Language, aimed at workloads needing Cassandra-style scale without operating nodes.

A managed service that moves data between homogeneous or heterogeneous sources and targets, supporting one-time loads and continuous replication with minimal downtime.

A desktop utility that converts database schemas, code objects, and stored procedures from one engine to another and flags objects that need manual rewriting.

Moving a database between different engines, such as Oracle to PostgreSQL, which typically requires schema and code conversion in addition to moving the data.

A ruggedized edge computing and storage device used to move very large datasets into or out of AWS physically when network transfer would be too slow.

A technique that keeps two identical environments, sends traffic to one, and switches to the other after validation, allowing fast rollback if problems appear.

An Aurora cluster replicated to secondary AWS Regions over dedicated infrastructure, enabling low-latency regional reads and cross-Region disaster recovery.

A managed connection pooler placed between applications and a relational database that absorbs connection churn and shortens failover time.

A container of engine configuration values applied to database instances, letting you tune settings and apply changes with or without a reboot.

A centralized managed service for defining backup policies and automating snapshot and point-in-time backups across supported AWS database and storage services.

Restoring a database to any second inside a retention window using continuous backup logs, usually by creating a new instance rather than overwriting the existing one.

A database tuning tool that visualizes workload by wait events, showing which SQL statements and resources consume the most database load.

An RDS feature that gathers operating-system metrics such as CPU, memory, and disk input/output at granular intervals from an agent running on the instance.

An Aurora MySQL feature that rewinds a cluster in place to an earlier point in time without restoring from backup, giving fast recovery from accidental changes.

An in-memory cache for DynamoDB that serves eventually consistent reads with microsecond latency and lowers read pressure on the underlying tables.

An encryption key that you create and control in AWS KMS, used to encrypt database storage and snapshots and to govern who may decrypt them.

A method of connecting to a relational database with a short-lived token generated from IAM credentials instead of a long-term stored database password.

An automated process in which AWS Secrets Manager periodically generates a new database credential and updates both the stored secret and the database user.

A stateful virtual firewall that controls inbound and outbound traffic to database resources, with rules referencing CIDR ranges or other security groups.

A collection of subnets inside a VPC that a managed relational database uses to place instances, determining which Availability Zones and IP ranges are available.

A volume type for databases where you specify input/output operations per second independently of volume size to meet consistent low-latency performance needs.