# Amazon Aurora

Cloud service · AWS. Updated 2026-11-12. Page: https://blowstack.com/tools/cloud-services-hub/amazon-aurora

> Amazon Aurora is a fully managed relational database service by Amazon Web Services (AWS) that is designed to be compatible with MySQL and PostgreSQL while offering the performance and availability of commercial databases with improved scalability, durability, and security.

Practice exams covering Amazon Aurora: [AIF-C01 AWS Certified AI Practitioner](https://blowstack.com/practice-exams/aws-certified-ai-practitioner), [CLF-C02 AWS Certified Cloud Practitioner](https://blowstack.com/practice-exams/aws-certified-cloud-practitioner), [SOA-C03 AWS Certified CloudOps Engineer - Associate](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate), [DEA-C01 AWS Certified Data Engineer - Associate](https://blowstack.com/practice-exams/aws-certified-data-engineer-associate), [DVA-C02 AWS Certified Developer - Associate](https://blowstack.com/practice-exams/aws-certified-developer-associate), [DOP-C02 AWS Certified DevOps Engineer - Professional](https://blowstack.com/practice-exams/aws-certified-devops-engineer-professional), [AIP-C01 AWS Certified Generative AI Developer - Professional](https://blowstack.com/practice-exams/aws-certified-generative-ai-developer-professional), [MLA-C02 (beta) AWS Certified Machine Learning Engineer - Associate](https://blowstack.com/practice-exams/aws-certified-machine-learning-engineer-associate), [SAA-C03 AWS Certified Solutions Architect - Associate](https://blowstack.com/practice-exams/aws-certified-solutions-architect-associate), [SAP-C03 AWS Certified Solutions Architect - Professional](https://blowstack.com/practice-exams/aws-certified-solutions-architect-professional).

## Free Amazon Aurora tools

- [RDS Max Connections Calculator](https://blowstack.com/tools/aws-rds-max-connections-calculator): Find Aurora MySQL's and Aurora PostgreSQL's default max_connections per instance class, and budget connections.

## 8 key features of Amazon Aurora

Amazon Aurora is a highly scalable, durable, and secure fully managed relational database with MySQL and PostgreSQL compatibility, offering superior performance, automated backups, easy monitoring, and recovery features.

1. **High Performance and Scalability**: Amazon Aurora offers six times the throughput of PostgreSQL and MySQL and automatically scales to workload demand. It is designed for up to 99.99% single-Region and 99.999% multi-Region availability.
2. **Fully Managed**: As a fully managed service, Aurora handles time-consuming tasks such as provisioning, patching, backup, recovery, failure detection, and repair. You also benefit from the integration with AWS services such as AWS Lambda and Amazon RDS.
3. **Compatibility with MySQL and PostgreSQL**: Amazon Aurora is fully compatible with existing MySQL and PostgreSQL open source databases, making it easy for database developers to transition applications without having to re-engineer their entire database.
4. **Security**: Aurora includes multiple layers of security, including network isolation using Amazon VPC, encryption at rest using keys you create and control through AWS Key Management Service (KMS), and encryption of data in transit using SSL. You also have the option to use IAM to define who can access the RDS databases.
5. **Durability and Availability**: Aurora is designed for up to 99.99% single-Region and 99.999% multi-Region availability. It also provides fault-tolerant distributed storage to ensure continuous data access and durability.
6. **Automated Backups, Database Snapshots, and Point-in-Time Recovery**: Continuous backups to Amazon S3 and automated snapshots enable point-in-time recovery. You can also easily create database snapshots, which allow you to quickly replicate databases across regions or accounts.
7. **Monitoring and Management**: Aurora provides an advanced monitoring system allowing you to track over 70 metrics through the AWS Management Console. Enhanced RDS monitoring provides visibility into your databases' performance and allows you to set alarms and automate actions based on specific conditions.
8. **Migration from RDS MySQL and PostgreSQL**: Bring data from RDS for MySQL or RDS for PostgreSQL into Aurora by creating and restoring snapshots or by setting up one-way replication.

## Where Amazon Aurora is used

Amazon Aurora is employed for hosting web and SaaS applications, analytics, e-commerce systems, gaming databases, legacy database migrations, and disaster recovery, highlighting its high performance, scalability, and availability.

1. **Web Applications Hosting**: Amazon Aurora serves as a scalable and reliable database engine for web applications, handling multiple concurrent user requests and large amounts of data with ease, while maintaining high performance and availability.
2. **Enterprise SaaS Applications**: Designed for software-as-a-service (SaaS) providers, Amazon Aurora allows for multi-tenancy architecture support while offering each tenant isolated, secure, and compliant data environments, crucial for enterprise-level SaaS solutions.
3. **High-Throughput Analytics**: Amazon Aurora is optimized for scenarios requiring complex queries and high-throughput analytics, with features like parallel query processing, which significantly speed up data analysis tasks.
4. **E-commerce Systems**: E-commerce platforms leverage Aurora for its high transactional throughput, automatic scaling, and data durability, ensuring that spikes in shopping traffic do not compromise database performance or data integrity.
5. **Gaming Databases**: Gaming platforms utilize Aurora to manage user data, support real-time leaderboards, and ensure high availability and fast access during peak gaming times, thereby enhancing player experience and engagement.
6. **Migration of Legacy Databases**: Aurora provides a seamless migration path for legacy databases, allowing businesses to move away from traditional, costly database solutions to a more cost-effective, managed database service without sacrificing features or performance.
7. **Disaster Recovery**: By utilizing Aurora's cross-region replication and automated backups, organizations can implement robust disaster recovery plans, ensuring that data remains safe and accessible even after system outages or regional disruptions.

## How Amazon Aurora is priced

Amazon Aurora offers pricing models including On-Demand Instances, Reserved Instances, Aurora Serverless with pay-per-use, and charges for additional storage and I/O usage.

- **On-Demand Instances**: Amazon Aurora on-demand instances charge by the hour with no long-term commitments or upfront fees. You pay for the compute capacity by the hour your instance is running.
- **Reserved Instances**: Reserved Instances offer discounts of up to 45% for a one-year term and up to 66% for a three-year term compared with On-Demand Instances.
- **Aurora Serverless**: Aurora Serverless pricing is based on the actual amount of resources consumed by your database, offering a variable pricing model that scales with your workloads automatically.
- **Additional Storage and I/O Costs**: Storage is billed per GB-month. Aurora Standard additionally charges for read and write I/O in increments of one million requests; Aurora I/O-Optimized has no read or write I/O operation charges.
- **Database Savings Plans**: Eligible Aurora usage can receive savings in exchange for a commitment to a specified amount of usage, measured in dollars per hour, over a one-year term.
- **Aurora DSQL**: Aurora DSQL uses a separate serverless pricing model described on the Aurora DSQL pricing page.
- **Aurora PostgreSQL Limitless Database**: Limitless Database compute capacity is measured in Aurora Capacity Units (ACUs) and billed per second; one ACU includes approximately 2 GiB of memory with corresponding CPU and networking.
- **CPU Credits**: T4g and T3 instances run in Unlimited mode and incur CPU-credit charges when average CPU utilization over a rolling 24-hour period exceeds the instance baseline. Credits cost $0.09 per vCPU-hour across instance sizes and Regions for both cluster configurations and are not covered by Reserved Instances.
- **Aurora Global Database**: Global Database charges for replicated write I/O between the primary Region and each secondary Region under both Aurora Standard and Aurora I/O-Optimized configurations.
- **Backtrack**: Backtrack charges an hourly rate for storing change records retained for the configured backtrack duration.
- **Backup Storage**: Backup storage up to 100% of the database cluster size and snapshots created within the backup retention period are free. Other backups and snapshots, including those from deleted clusters, are metered per GB-month.
- **Data API**: Data API charges for API and data requests without minimum fees or upfront commitments. Request payloads are metered in 32 KB increments, with an additional billable request for each increment.
- **Data Transfer**: Aurora data transfer with EC2 in the same Availability Zone and cross-AZ DB cluster replication are free. Transfers between EC2 and Aurora instances in different Availability Zones of the same Region incur EC2 Regional Data Transfer charges.
- **RDS Extended Support**: RDS Extended Support is charged per vCPU-hour for provisioned Aurora instances and per consumed ACU-hour for Aurora serverless. Rates depend on AWS Region and calendar date.
- **Snapshot and Cluster Export**: Snapshot and cluster exports are metered per gigabyte of snapshot data. Exporting only part of a cluster still incurs charges for the entire cluster, and repeated exports from the same snapshot are not incremental.
- **Zero-ETL Integrations**: Zero-ETL integrations have no additional integration fee, but resources used to create and process their change data can incur charges.
- **AWS Free Tier**: New AWS customers receive $100 in credits at sign-up and can earn another $100, usable across eligible services including Aurora for up to 12 months. The Free plan supports Aurora PostgreSQL serverless instances with up to 4 ACUs and 1 GiB of storage per cluster.
- **On-Demand Capacity Reservations**: Reserve DB instance capacity in a specific Availability Zone for as long as needed. Reserved capacity is charged at the On-Demand rate whether used or unused; when used, you pay for the instance without a separate reservation charge. There are no upfront or additional charges.

## Services Amazon Aurora works with

- **[Amazon RDS](https://blowstack.com/tools/cloud-services-hub/amazon-rds)**: Amazon Aurora is a fully managed relational database engine, built within the Amazon RDS ecosystem, and provides MySQL and PostgreSQL compatible database engines.
- **[AWS Lambda](https://blowstack.com/tools/cloud-services-hub/aws-lambda)**: You can use AWS Lambda to trigger functions in response to events in Amazon Aurora, allowing for serverless computing within your database environment.
- **[Amazon S3](https://blowstack.com/tools/cloud-services-hub/amazon-s3)**: Amazon Aurora integrates with Amazon S3 for importing and exporting data, allowing for easy data migration and backups.
- **[Amazon CloudWatch](https://blowstack.com/tools/cloud-services-hub/amazon-cloudwatch)**: Integration with Amazon CloudWatch allows you to monitor Amazon Aurora database instances for performance and operational health.
- **[AWS CloudFormation](https://blowstack.com/tools/cloud-services-hub/aws-cloudformation)**: You can use AWS CloudFormation to deploy and manage Amazon Aurora databases as part of infrastructure as code.
- **[Amazon Athena](https://blowstack.com/tools/cloud-services-hub/amazon-athena)**: Aurora databases can be queried from Athena through federated query connectors for MySQL and PostgreSQL. This lets you join Aurora records with data in Amazon S3 without first copying the database into a data lake.
- **[Amazon Aurora Serverless](https://blowstack.com/tools/cloud-services-hub/amazon-aurora-serverless)**: Aurora supports serverless database capacity that adjusts to application demand. Aurora Serverless v2 can provide automatically scaling writer and reader instances within an Aurora cluster.
- **[Amazon Bedrock](https://blowstack.com/tools/cloud-services-hub/amazon-bedrock)**: Aurora PostgreSQL can store embeddings and source metadata for Amazon Bedrock Knowledge Bases using the pgvector extension. Bedrock retrieves relevant records from Aurora to ground generated answers in your documents.
- **[Amazon Comprehend](https://blowstack.com/tools/cloud-services-hub/amazon-comprehend)**: Aurora machine learning lets SQL queries call Amazon Comprehend for natural-language analysis. For example, a query can classify the sentiment of customer feedback stored in an Aurora table.
- **[Amazon Data Firehose](https://blowstack.com/tools/cloud-services-hub/amazon-kinesis-data-firehose)**: Aurora can serve as a database source for Amazon Data Firehose, which captures database changes and delivers them to configured destinations. This supports continuously moving transactional data into analytics storage without maintaining a custom ingestion application.
- **[Amazon DevOps Guru](https://blowstack.com/tools/cloud-services-hub/amazon-devops-guru)**: Aurora integrates with DevOps Guru for RDS to detect database performance anomalies. The service analyzes database load and related metrics and provides findings that help identify causes such as contention or problematic SQL.
- **[Amazon EC2](https://blowstack.com/tools/cloud-services-hub/amazon-ec2)**: Aurora provides a connection setup feature that configures security groups so a selected EC2 instance can reach a database. This simplifies establishing database connectivity for an application hosted on EC2.
- **[Amazon EMR](https://blowstack.com/tools/cloud-services-hub/amazon-emr)**: Aurora tables can be read and written by processing jobs on Amazon EMR through JDBC connectors. For example, an EMR Spark job can extract Aurora records for large-scale transformations and write aggregated results back.
- **[Amazon EventBridge](https://blowstack.com/tools/cloud-services-hub/amazon-eventbridge)**: Aurora publishes database events to EventBridge, including cluster and instance lifecycle events. EventBridge rules can route matching events to automation or notification targets, such as invoking a function after a failover.
- **[Amazon GuardDuty](https://blowstack.com/tools/cloud-services-hub/amazon-guardduty)**: Aurora integrates with GuardDuty RDS Protection to detect suspicious database login activity. GuardDuty analyzes supported Aurora login events and produces findings for behavior such as anomalous access or potential credential misuse.
- **[Amazon Kendra](https://blowstack.com/tools/cloud-services-hub/amazon-kendra)**: Aurora can supply database content to Amazon Kendra through its database connector. Kendra indexes selected records so users can search database-backed content alongside other enterprise information.
- **[Amazon Kinesis Data Streams](https://blowstack.com/tools/cloud-services-hub/amazon-kinesis-data-streams)**: Aurora Database Activity Streams publishes database activity to a Kinesis data stream. Audit and security consumers can read the stream to inspect database activity outside the database itself.
- **[Amazon Managed Grafana](https://blowstack.com/tools/cloud-services-hub/amazon-managed-grafana)**: Aurora databases can be connected to Amazon Managed Grafana using its MySQL or PostgreSQL data sources. Grafana dashboards can run SQL queries against Aurora to visualize application and business data.
- **[Amazon Managed Service for Apache Flink](https://blowstack.com/tools/cloud-services-hub/amazon-managed-service-for-apache-flink)**: Aurora can act as a relational source or destination for Managed Service for Apache Flink applications using compatible JDBC connectors. A streaming application can, for example, write continuously calculated aggregates into Aurora tables.
- **[Amazon Managed Workflows for Apache Airflow (Amazon MWAA)](https://blowstack.com/tools/cloud-services-hub/amazon-managed-workflows-for-apache-airflow-(amazon-mwaa))**: Aurora SQL operations can be orchestrated from Amazon MWAA using Airflow's RDS Data API integration. Workflows can execute statements against supported Aurora databases as part of scheduled data-processing pipelines.
- **[Amazon OpenSearch Service](https://blowstack.com/tools/cloud-services-hub/amazon-opensearch-service)**: Aurora can replicate database records into Amazon OpenSearch Service through supported OpenSearch Ingestion database integrations. Initial loading and ongoing change capture keep search indexes synchronized with transactional tables.
- **[Amazon Q Business](https://blowstack.com/tools/cloud-services-hub/amazon-q-business)**: Aurora databases can be indexed by Amazon Q Business through supported database connectors. Selected database content becomes available for retrieval when Q Business answers employee questions.
- **[Amazon QuickSight](https://blowstack.com/tools/cloud-services-hub/amazon-quicksight)**: Aurora databases are supported data sources for Amazon QuickSight. You can query Aurora directly or import records into SPICE to build dashboards and analyze transactional data.
- **[Amazon RDS Proxy](https://blowstack.com/tools/cloud-services-hub/amazon-rds-proxy)**: Aurora can use RDS Proxy to pool and reuse database connections and improve application resilience during database disruptions. Applications connect to a proxy endpoint, which manages connections to the registered Aurora cluster.
- **[Amazon Redshift](https://blowstack.com/tools/cloud-services-hub/amazon-redshift)**: Aurora supports zero-ETL integrations that replicate transactional data into Amazon Redshift for analytics. Redshift also supports federated queries against supported Aurora databases, allowing warehouse queries to access operational tables.
- **[Amazon SageMaker](https://blowstack.com/tools/cloud-services-hub/amazon-sagemaker)**: Aurora machine learning can invoke SageMaker inference endpoints directly from SQL to generate predictions for database records. Supported Aurora zero-ETL integrations can also make transactional data available in the SageMaker lakehouse.
- **[Amazon Simple Notification Service (Amazon SNS)](https://blowstack.com/tools/cloud-services-hub/amazon-simple-notification-service-(amazon-sns))**: Aurora event subscriptions deliver selected database events to SNS topics. Subscribers can receive notifications about events such as failovers, maintenance, and configuration changes.
- **[Amazon VPC](https://blowstack.com/tools/cloud-services-hub/amazon-vpc)**: Aurora clusters use VPC subnets and security groups to control database network placement and access. DB subnet groups identify the subnets available for database instances across Availability Zones.
- **[AWS Amplify](https://blowstack.com/tools/cloud-services-hub/aws-amplify)**: Aurora MySQL and PostgreSQL databases can be connected to an Amplify backend using Amplify's existing SQL database integration. Amplify can generate an API and data models from the database schema for application access.
- **[AWS AppSync](https://blowstack.com/tools/cloud-services-hub/aws-appsync)**: Aurora databases that support the RDS Data API can serve as relational data sources for AWS AppSync. GraphQL resolvers execute SQL statements against Aurora to retrieve or update application data.
- **[AWS Auto Scaling](https://blowstack.com/tools/cloud-services-hub/aws-auto-scaling)**: Aurora Auto Scaling uses Application Auto Scaling to adjust the number of Aurora Replicas in a cluster. Target-tracking policies can add or remove readers based on metrics such as average reader CPU utilization.
- **[AWS Backup](https://blowstack.com/tools/cloud-services-hub/aws-backup)**: Aurora clusters can be protected through AWS Backup plans. AWS Backup manages scheduled recovery points, retention, supported copies, and restores alongside backups of other AWS resources.
- **[AWS CloudTrail](https://blowstack.com/tools/cloud-services-hub/aws-cloudtrail)**: Aurora management API activity is recorded in CloudTrail. These records identify actions such as creating clusters, changing configurations, or deleting database resources, along with the identity that performed them.
- **[AWS Config](https://blowstack.com/tools/cloud-services-hub/aws-config)**: Aurora cluster and instance configurations can be recorded and evaluated by AWS Config. Rules can check settings such as storage encryption, public accessibility, and deletion protection.
- **[AWS Control Tower](https://blowstack.com/tools/cloud-services-hub/aws-control-tower)**: Aurora resources can be governed by applicable AWS Control Tower controls for Amazon RDS. Proactive controls can reject noncompliant CloudFormation deployments, including database clusters that lack required security settings.
- **[AWS Data Pipeline](https://blowstack.com/tools/cloud-services-hub/aws-data-pipeline)**: Aurora MySQL can be accessed by AWS Data Pipeline through its relational database and SQL activity capabilities. Pipelines can run SQL or copy database records to destinations such as Amazon S3.
- **[AWS Database Migration Service (AWS DMS)](https://blowstack.com/tools/cloud-services-hub/aws-database-migration-service-(aws-dms))**: Aurora can be a source or target for AWS DMS migrations using supported MySQL- or PostgreSQL-compatible endpoints. DMS can perform an initial data load and replicate ongoing changes to reduce migration downtime.
- **[AWS Directory Service](https://blowstack.com/tools/cloud-services-hub/aws-directory-service)**: Aurora PostgreSQL can use AWS Managed Microsoft AD for Kerberos authentication. Database users can authenticate with directory identities instead of maintaining separate database passwords.
- **[AWS Fault Injection Service](https://blowstack.com/tools/cloud-services-hub/aws-fault-injection-simulator-(aws-fis))**: Aurora clusters support fault-injection experiments through AWS Fault Injection Service actions for RDS. Experiments can trigger a cluster failover or reboot database instances to test application recovery behavior.
- **[AWS Glue](https://blowstack.com/tools/cloud-services-hub/aws-glue)**: Aurora can be crawled and accessed by AWS Glue using JDBC connections. Glue jobs can extract and transform Aurora data or write processed results back into database tables.
- **[AWS Glue DataBrew](https://blowstack.com/tools/cloud-services-hub/aws-glue-databrew)**: Aurora databases can supply datasets to AWS Glue DataBrew through supported JDBC connections. DataBrew lets users profile and prepare database data with visual transformation recipes.
- **[AWS Health Dashboard](https://blowstack.com/tools/cloud-services-hub/aws-health-dashboard)**: Aurora resource health and planned service changes can appear in AWS Health Dashboard. Account-specific events identify affected resources and provide information about issues or required maintenance actions.
- **[AWS Identity and Access Management (IAM)](https://blowstack.com/tools/cloud-services-hub/aws-identity-and-access-management-(iam))**: Aurora uses IAM policies to authorize management operations and IAM roles to access integrated AWS services. Supported database engines also offer IAM database authentication, allowing clients to connect using short-lived authentication tokens.
- **[AWS Key Management Service (AWS KMS)](https://blowstack.com/tools/cloud-services-hub/aws-key-management-service-(aws-kms))**: Aurora uses KMS keys to encrypt database storage, backups, and snapshots. KMS also protects the encryption keys used by Aurora Database Activity Streams.
- **[AWS PrivateLink](https://blowstack.com/tools/cloud-services-hub/aws-privatelink)**: Aurora management operations and supported RDS Data API requests can use interface VPC endpoints powered by AWS PrivateLink. This keeps API traffic on private connectivity without requiring an internet gateway or NAT gateway.
- **[AWS Resilience Hub](https://blowstack.com/tools/cloud-services-hub/aws-resilience-hub)**: Aurora resources can be assessed by AWS Resilience Hub as part of an application's resilience evaluation. The assessment considers database deployment and recovery arrangements when evaluating recovery objectives and recommending improvements.
- **[AWS Resource Access Manager (AWS RAM)](https://blowstack.com/tools/cloud-services-hub/aws-resource-access-manager-(aws-ram))**: Aurora uses AWS RAM to share supported DB clusters with other AWS accounts for cross-account cloning. A recipient can create an Aurora clone without receiving access to the source cluster's database endpoint.
- **[AWS Resource Groups and Tag Editor](https://blowstack.com/tools/cloud-services-hub/aws-resource-groups-and-tag-editor)**: Aurora clusters and database instances can be organized into tag-based resource groups. Tag Editor can discover supported Aurora resources and apply or update their tags in bulk.
- **[AWS Secrets Manager](https://blowstack.com/tools/cloud-services-hub/aws-secrets-manager)**: Aurora can manage its master-user credentials in Secrets Manager, including password rotation. Secrets Manager also supports storing and rotating credentials for other Aurora database users.
- **[AWS Security Hub](https://blowstack.com/tools/cloud-services-hub/aws-security-hub)**: Aurora resources are evaluated by applicable RDS security controls in AWS Security Hub. Findings can identify issues such as unencrypted clusters, missing deletion protection, or insufficient audit logging.
- **[AWS Step Functions](https://blowstack.com/tools/cloud-services-hub/aws-step-functions)**: Aurora operations can be orchestrated with Step Functions service integrations for Amazon RDS and the RDS Data API. A workflow can execute SQL against a supported cluster or automate database lifecycle operations with retries and error handling.
- **[AWS Systems Manager](https://blowstack.com/tools/cloud-services-hub/aws-systems-manager)**: Aurora maintenance and lifecycle operations can be automated with Systems Manager Automation runbooks. Runbooks can coordinate actions such as starting or stopping supported clusters and collecting diagnostic information.
- **[AWS Trusted Advisor](https://blowstack.com/tools/cloud-services-hub/aws-trusted-advisor)**: Aurora configurations are covered by applicable Trusted Advisor checks, including checks related to cluster fault tolerance. Recommendations can highlight deployment choices that leave a cluster without adequate failover capacity.
- **[AWS User Notifications](https://blowstack.com/tools/cloud-services-hub/aws-user-notifications)**: Aurora events delivered through EventBridge can be selected in AWS User Notifications configurations. Matching cluster or instance events can generate notifications through configured delivery channels.
- **[AWS X-Ray](https://blowstack.com/tools/cloud-services-hub/aws-x-ray)**: Aurora database calls can be traced through X-Ray instrumentation for supported MySQL and PostgreSQL clients. Traces show database request timing and errors alongside the rest of an application's request path.

## Amazon Aurora exam questions

Real practice exam questions on Amazon Aurora, with the correct answer and its explanation.

### Question 1 (AIF-C01)

**Question:**

A company is building a generative AI assistant that retrieves relevant product documentation before answering customer questions. The application uses PostgreSQL drivers and SQL joins to combine documentation with product and customer records. The company wants to store document embeddings alongside these relational records in one managed database and retrieve documents by semantic similarity without rewriting the existing PostgreSQL queries. Which solution best meets these requirements?

**Correct answer:** Use Amazon Aurora PostgreSQL-Compatible Edition with the pgvector extension to store embeddings and perform vector similarity searches alongside relational queries.

**Explanation:** Aurora PostgreSQL supports the pgvector extension for storing and searching embeddings. It can keep embeddings and relational records in the same managed database while supporting PostgreSQL drivers and SQL joins.

### Question 2 (MLA-C02 (beta))

**Question:**

A distributor is building a generative AI assistant for sales analysts. Its source data consists of millions of relational records for orders, order items, products, and refunds. Analysts ask questions such as, "What was net revenue for industrial pumps in the western region last quarter, after subtracting refunds?"

The assistant must calculate exact results across all qualifying records, support different combinations of product, region, and date filters, and explain the results using retrieved evidence. The company wants to avoid maintaining a separate search index or precomputing summaries for every possible filter combination.

Which retrieval architecture BEST meets these requirements?

**Correct answer:** Store the records in Amazon Aurora PostgreSQL-Compatible Edition. Map each question to a validated, parameterized, read-only SQL query that joins and aggregates the required tables. Supply the query results and calculation metadata to the foundation model as grounding context.

**Explanation:** Structured-data RAG can retrieve evidence through SQL rather than vector search. Aurora can perform joins, filtering, and exact aggregation across all qualifying records. Validated query templates and parameters constrain database access, while the returned totals, filters, and calculation definitions ground the generated explanation.

### Question 3 (SAP-C03)

**Question:**

A manufacturer built a production-planning prototype in Amazon Redshift to analyze completed factory jobs. The historical analysis has since moved to a separate platform. The remaining database will support an interactive application used by thousands of factory operators.

Each application request retrieves a few job records by identifier or atomically updates a job assignment, a machine reservation, and a material balance. Requests must complete within 100 milliseconds at the expected peak load. The application also executes SQL joins across current jobs, machines, and materials. Its existing relational schema and SQL transaction logic must be retained.

Load tests show that large historical scans are no longer significant, but frequent small updates and concurrent point lookups do not meet the latency requirement. The company wants a managed database architecture suited to this new workload, without application-managed sharding.

Which solution should a solutions architect recommend?

**Correct answer:** Migrate the operational tables to Amazon Aurora PostgreSQL-Compatible Edition. Adapt engine-specific SQL as needed while retaining the relational schema and transaction logic, index the application's lookup and join columns, and execute the short transactions on the writer.

**Explanation:** The remaining workload is OLTP: concurrent indexed lookups and small, atomic updates across related records. Aurora provides a managed relational engine suited to these operations and preserves relational joins and multi-table transactions. Appropriate indexes and writer sizing should be validated against the latency target. A separate analytical copy is unnecessary because historical analysis has already moved elsewhere.

### Question 4 (CLF-C02)

**Question:**

A company is moving a PostgreSQL application to AWS. The application experiences unpredictable changes in both read and write traffic. The company wants a managed relational database that automatically adjusts compute capacity to match demand without requiring administrators to select new DB instance sizes. Which solution best meets these requirements?

**Correct answer:** Amazon Aurora PostgreSQL-Compatible with Aurora Serverless v2

**Explanation:** Aurora Serverless v2 automatically adjusts database compute capacity within configured limits as demand changes. It supports PostgreSQL-compatible applications without requiring administrators to change DB instance sizes.

### Question 5 (SOA-C03)

**Question:**

A company runs a read-heavy application on a provisioned Amazon Aurora MySQL cluster with one writer and two Aurora Replicas. Read-only connections use the reader endpoint, and connection pools regularly recycle connections. Traffic increases unpredictably.

Load testing shows that each reader meets the application's latency requirement with up to 100 concurrent database connections. Above that level, latency increases even though reader CPU utilization remains below 30%. Almost all connections are actively executing queries, and the database connection limit is not being reached. The writer has ample capacity.

A CloudOps engineer must automatically add and remove readers as connection demand changes, maintaining between two and eight readers. Which configuration meets these requirements?

**Correct answer:** Configure Aurora Auto Scaling with two to eight replicas and a target tracking policy that targets 100 average database connections per reader.

**Explanation:** Aurora Auto Scaling supports target tracking with the predefined RDSReaderAverageDatabaseConnections metric. A target of 100 aligns scaling with the tested per-reader workload limit. The policy can add readers as average connection demand increases and remove them as demand falls, within the configured limits.

### Question 6 (AIP-C01)

**Question:**

A company uses Amazon Aurora PostgreSQL-Compatible Edition with pgvector to retrieve context for a generative AI application. A document table has an HNSW index using vector_cosine_ops. Retrieval queries use the matching cosine-distance operator, and EXPLAIN ANALYZE confirms that they use the index.

Most requests use a fast retrieval tier. A new investigation tier requires recall@20 of at least 97% compared with exact nearest-neighbor results. Representative benchmarks show that hnsw.ef_search = 40 provides 88% recall with 45-ms p95 retrieval latency, while hnsw.ef_search = 160 provides 98% recall with 115-ms p95 retrieval latency. The investigation tier has a 150-ms retrieval budget. Fast-tier requests must retain their existing search settings and performance.

The application reuses database sessions through a connection pool. Both request types execute the same parameterized retrieval query and return 20 chunks. The developer must apply the investigation-tier tuning without allowing it to affect subsequent fast-tier requests that reuse a connection.

Which solution best meets these requirements?

**Correct answer:** For investigation requests, begin an explicit transaction, execute SET LOCAL hnsw.ef_search = 160, run the retrieval query on that same connection, and commit. Roll back the transaction on failure before returning the connection to the pool.

**Explanation:** hnsw.ef_search controls the query-time search candidate list, trading additional search work for improved recall. The benchmark establishes that 160 meets both investigation-tier requirements. SET LOCAL scopes the change to the current transaction, so committing or rolling back restores the connection's previous setting before another request reuses it.

### Question 7 (DEA-C01)

**Question:**

A financial research company stores document embeddings in Amazon Aurora PostgreSQL-Compatible Edition with pgvector. Its current vector index searches a multilayer proximity graph and meets the latency and recall targets for an interactive research application. Testing shows that evaluating distances against every stored embedding exceeds the interactive application's latency limit.

The company adds a nightly audit workload. For each audit query, the workload must return the actual 10 closest embeddings across the entire corpus under the configured distance measure. Occasionally omitting a closer embedding is unacceptable. Audit queries can run longer, but the interactive application must retain its current performance.

Which approach meets both workloads' requirements?

**Correct answer:** Keep the HNSW index for interactive searches. For audit queries, use an exact search plan that evaluates every stored embedding and returns the 10 smallest distances.

**Explanation:** HNSW provides approximate nearest-neighbor retrieval and is appropriate for the interactive workload's latency and recall requirements. A separate exhaustive exact-search path guarantees the audit's nearest-neighbor results without replacing the interactive application's index.

### Question 8 (DOP-C02)

**Question:**

A company runs a transaction-processing application on an Amazon Aurora PostgreSQL global database with one primary DB cluster and one secondary DB cluster in another AWS Region. Both clusters have provisioned DB instances and use an engine version that supports managed global database failover and recovery point objective enforcement.

During a network impairment, the primary cluster continues acknowledging transactions while cross-Region replication falls several minutes behind. The company requires the database to enforce a maximum cross-Region recovery point objective (RPO) of 60 seconds. If replication cannot maintain this objective, delaying transaction commits in the primary Region is acceptable. Regional recovery will use a managed failover to the existing secondary cluster. The company does not want custom application write-gating logic.

Which change should a DevOps engineer implement?

**Correct answer:** Configure the primary cluster's rds.global_db_rpo parameter to 60 seconds in its DB cluster parameter group.

**Explanation:** Aurora PostgreSQL global databases support RPO enforcement through rds.global_db_rpo. With one secondary Region, Aurora can block primary transaction commits when necessary until that secondary catches up sufficiently to satisfy the configured RPO. This provides database-level enforcement and explicitly trades primary write availability for the required data protection.

### Question 9 (DVA-C02)

**Question:**

A developer maintains a document-editing application that uses Amazon Aurora PostgreSQL. Each document row contains document_id, content, and an integer version. When an editor opens a document, the application returns its content and version. Editors might keep documents open for several minutes. If another editor saves a change during that time, saving the older form must return a conflict instead of overwriting the newer content. The application must not hold database locks while editors work. Which implementation meets these requirements?

**Correct answer:** When saving, run UPDATE documents SET content = :content, version = version + 1 WHERE document_id = :id AND version = :submitted_version. Return a conflict if no row was updated.

**Explanation:** The conditional UPDATE atomically checks the submitted version and applies the change. Incrementing the version invalidates other forms based on the previous version. If a concurrent save has already changed the version, no row is updated and the application can return a conflict without holding locks during editing.

### Question 10 (SAA-C03)

**Question:**

A retail company is moving a MySQL reporting application to AWS. The application runs aggregate queries that scan billions of sales records. The dataset is much larger than the database instance's memory, and loading scanned data into the buffer pool provides little benefit because successive reports access different records.

The company wants a managed, MySQL-compatible database that can execute eligible scan, filtering, and partial aggregation operations in its distributed storage layer instead of performing all this work on database instance compute. The application must continue querying current operational tables without maintaining a separate analytics dataset. Compatibility testing confirms that the reporting queries are eligible for storage-layer execution.

Which solution meets these requirements?

**Correct answer:** Use Amazon Aurora MySQL-Compatible on a supported engine version and DB instance class, and enable Aurora Parallel Query.

**Explanation:** Aurora Parallel Query pushes eligible query operations, including scans, filtering, and partial aggregation, into Aurora's distributed storage layer. It can accelerate large analytical queries without requiring the scanned dataset to pass through the instance's buffer pool or maintaining a separate analytics copy.

## Practice Amazon Aurora in a full exam

Questions on Amazon Aurora in each practice exam, with every answer option and an explanation for each:

- [SAA-C03](https://blowstack.com/practice-exams/aws-certified-solutions-architect-associate): 29 questions
- [SAP-C03](https://blowstack.com/practice-exams/aws-certified-solutions-architect-professional): 17 questions
- [SOA-C03](https://blowstack.com/practice-exams/aws-certified-cloudops-engineer-associate): 14 questions
- [DEA-C01](https://blowstack.com/practice-exams/aws-certified-data-engineer-associate): 12 questions
- [AIP-C01](https://blowstack.com/practice-exams/aws-certified-generative-ai-developer-professional): 8 questions
- [MLA-C02 (beta)](https://blowstack.com/practice-exams/aws-certified-machine-learning-engineer-associate): 5 questions
- [DOP-C02](https://blowstack.com/practice-exams/aws-certified-devops-engineer-professional): 5 questions
- [CLF-C02](https://blowstack.com/practice-exams/aws-certified-cloud-practitioner): 4 questions
- [AIF-C01](https://blowstack.com/practice-exams/aws-certified-ai-practitioner): 2 questions
- [DVA-C02](https://blowstack.com/practice-exams/aws-certified-developer-associate): 2 questions

## What is Amazon Aurora

Amazon Aurora is a fully managed relational database engine that combines the speed and reliability of high-end commercial databases with the simplicity and cost-effectiveness of open-source databases.

Introduced by Amazon Web Services (AWS) in 2014, Aurora has quickly become one of the popular choices for companies looking to deploy databases in the cloud. At its core, Aurora is designed to be compatible with MySQL and PostgreSQL, which allows developers to use the same code, drivers, and tools that they are already familiar with. However, under the hood, Amazon has heavily optimized Aurora for the cloud, providing performance improvements while ensuring strong durability and reliability.

Aurora also offers Aurora DSQL, a serverless, distributed SQL database with PostgreSQL compatibility and active-active multi-Region availability.

One of the critical innovations of Amazon Aurora is its distributed, fault-tolerant, self-healing storage system, which automatically replicates data across multiple Availability Zones (AZs). Each Aurora database volume is backed by a fleet of storage instances that AWS manages across three AZs to ensure high availability and durability. The system is designed to handle the loss of up to two copies of data without affecting database write availability and up to three copies without affecting read availability.

Aurora PostgreSQL and Aurora MySQL offer Aurora Standard and Aurora I/O-Optimized cluster configurations. I/O-Optimized is intended for I/O-intensive applications that need improved price performance and predictable pricing.

Feature and option support varies by database engine version and AWS Region.

Aurora offers six times the throughput of PostgreSQL and MySQL. This boost is achieved through a combination of optimizations like advanced read-ahead and write-behind mechanisms, as well as the tight integration within the AWS ecosystem. Moreover, Amazon Aurora is highly scalable. The underlying storage automatically scales up to 256 TiB per database instance.

Aurora PostgreSQL and Aurora MySQL can run with provisioned instances, whose resources depend on the selected instance type, or with Aurora serverless, which automatically starts, shuts down, and adjusts capacity to application demand.

The engine supports up to 15 low-latency read replicas, which helps in scaling read operations and enhancing the performance of read-heavy applications. Amazon Aurora also simplifies database management tasks. Since it's a managed service, many of the time-consuming tasks such as provisioning, patching, backup, recovery, failure detection, and repair are automated.

An Aurora MySQL cluster has a single primary open for reads and writes, with read-only replicas that can be automatically promoted to primary after a failure. The primary sends redo log records over the network rather than data pages.

The RDS management API has default throttling limits per AWS account and Region: a token bucket capacity of 300 requests and a refill rate of 12 requests per second. Account-specific increases can raise these limits.

Aurora Global Database lets a single Aurora database span multiple Regions, providing low-latency local reads in each Region and disaster recovery from Region-wide outages.

RDS Extended Support lets Aurora MySQL and PostgreSQL databases continue using major versions after community end-of-life. AWS supplies critical security and bug fixes through patch releases for up to three additional years while you prepare an upgrade.

Aurora PostgreSQL Limitless Database provides automated horizontal scaling beyond a single instance's write-throughput and storage limits by distributing workloads over multiple writer instances while retaining a single database.

Aurora continuously backs up your data to Amazon S3, and these backups are incremental and continuous, meaning that you can quickly recover your database to any point within the saved history.

For Aurora MySQL, Backtrack can move a database to an earlier point in time without restoring a backup, helping recover from mistakes such as dropping a table or deleting the wrong row.

Snapshot export and cluster export write data from an Aurora snapshot or database cluster to Amazon S3 in Parquet format for analysis with services such as Amazon Athena, Amazon EMR, and Amazon SageMaker.

A comprehensive set of security features also comes built into Amazon Aurora. It integrates with AWS Identity and Access Management (IAM) and offers encryption at rest using AWS Key Management Service (KMS), as well as encryption in transit with SSL. Network isolation using Amazon VPC, along with the ability to create private endpoints, ensures that your data is protected within your own virtual network.

Data API provides a secure HTTPS API for executing SQL queries against Aurora databases, eliminating the network and application configuration tasks otherwise needed to establish a secure database connection.

Given these sophisticated features, Amazon Aurora is well-suited to a wide range of applications. Its high-performance characteristics make it ideal for enterprise applications, SaaS products, and any scenario that requires a robust, fault-tolerant, and scalable database. With Aurora, businesses and developers have the ability to support heavy workloads while benefiting from the ease of use and cost savings that come with a cloud-native database service.

## Learn Amazon Aurora in lessons

22 short lessons on Amazon Aurora in 9 certification courses.

- AIF-C01: Amazon Aurora: storing and searching embeddings (Premium), https://app.blowstack.com/lessons/amazon-aurora-storing-and-searching-embeddings-aif-c01
- MLA-C02 (beta): Amazon Aurora: vector retrieval with PostgreSQL (Premium), https://app.blowstack.com/lessons/amazon-aurora-vector-retrieval-with-postgresql-mla-c02
- SAP-C03: Amazon Aurora: cluster replication and failover (Premium), https://app.blowstack.com/lessons/amazon-aurora-cluster-replication-and-failover-sap-c03
- CLF-C02: Amazon Aurora: a managed relational database engine (Premium), https://app.blowstack.com/lessons/amazon-aurora-a-managed-relational-database-engine-clf-c02
- SOA-C03: Amazon Aurora and Amazon RDS: scaling differences (Premium), https://app.blowstack.com/lessons/amazon-aurora-and-amazon-rds-scaling-differences-soa-c03
- AIP-C01: Amazon Aurora: vector search with pgvector (Premium), https://app.blowstack.com/lessons/amazon-aurora-vector-search-with-pgvector-aip-c01

## Concepts related to Amazon Aurora

- [Vector Search](https://blowstack.com/tools/cloud-concepts-hub/vector-search): Vector search retrieves items by comparing numerical representations called vectors and ranking them by similarity to a query vector.
- [Database Replication](https://blowstack.com/tools/cloud-concepts-hub/database-replication): Database replication copies database data and propagates changes to other database instances or locations to support migration, availability, and read scaling.
- [Heterogeneous Migration](https://blowstack.com/tools/cloud-concepts-hub/heterogeneous-migration): Heterogeneous migration moves a database from one database engine to a different engine, requiring compatible data structures and application behavior on the destination.
- [Vertical Scaling](https://blowstack.com/tools/cloud-concepts-hub/vertical-scaling): Vertical scaling changes the capacity of an existing compute resource or database instance by increasing or decreasing its allocated resources rather than changing the number of instances.
- [Eventual Consistency](https://blowstack.com/tools/cloud-concepts-hub/eventual-consistency): Eventual consistency is a distributed-data model in which replicas may temporarily return different values but converge if updates stop and replication completes successfully.
- [Read/Write Splitting](https://blowstack.com/tools/cloud-concepts-hub/readwrite-splitting): Read/Write Splitting routes database writes to a primary database and eligible reads to replicas so that read workloads can scale independently.
- [Retrieval-Augmented Generation](https://blowstack.com/tools/cloud-concepts-hub/retrieval-augmented-generation): Retrieval-Augmented Generation is an application pattern that retrieves relevant external information and supplies it as context to a generative model when producing an answer.
- [Data Replication](https://blowstack.com/tools/cloud-concepts-hub/data-replication): Data replication is the process of maintaining copies of the same data on multiple storage systems or locations and propagating changes between them.
- [Point-in-Time Recovery](https://blowstack.com/tools/cloud-concepts-hub/point-in-time-recovery): Point-in-Time Recovery restores a database or other supported data store to its recoverable state at a selected time within a retained recovery window.
- [Recovery Point Objective](https://blowstack.com/tools/cloud-concepts-hub/recovery-point-objective): Recovery Point Objective (RPO) is the maximum acceptable amount of data loss from a disruption, expressed as a duration of time.
- [Relational Data Model](https://blowstack.com/tools/cloud-concepts-hub/relational-data-model): The relational data model organizes data into tables of rows and columns, using keys to identify records and represent relationships between tables.
- [Database](https://blowstack.com/tools/cloud-concepts-hub/database): A database is an organized collection of data that applications can efficiently store, retrieve, update, and manage.
- [Embeddings](https://blowstack.com/tools/cloud-concepts-hub/embeddings): Embeddings are learned numerical representations that encode characteristics of data as vectors so that meaningful relationships can be compared mathematically.
- [Hybrid Search](https://blowstack.com/tools/cloud-concepts-hub/hybrid-search): Hybrid search combines keyword-based retrieval and semantic vector retrieval to produce a single ranked set of relevant results.
- [ACID Transactions](https://blowstack.com/tools/cloud-concepts-hub/acid-transactions): ACID transactions group data operations into a reliable unit of work with atomicity, consistency, isolation, and durability guarantees.
- [Database Indexing](https://blowstack.com/tools/cloud-concepts-hub/database-indexing): Database indexing uses auxiliary data structures to locate records efficiently without scanning all stored data.
- [Deadlocks](https://blowstack.com/tools/cloud-concepts-hub/deadlocks): Deadlocks occur when transactions or processes form a cycle of resource dependencies, leaving each waiting for another to release a resource.
- [Hierarchical Indexing](https://blowstack.com/tools/cloud-concepts-hub/hierarchical-indexing): Hierarchical indexing organizes searchable data into multiple levels so queries can navigate from broad groups or coarse representations to more specific candidates.
- [Metadata Schemas](https://blowstack.com/tools/cloud-concepts-hub/metadata-schemas): Metadata schemas define the fields, data types, meanings, and validation rules used to describe documents and other resources consistently.
- [Polyglot Persistence](https://blowstack.com/tools/cloud-concepts-hub/polyglot-persistence): Polyglot persistence is the use of different database technologies within an application or system to match distinct data models, access patterns, and operational requirements.
- [Synchronous Replication](https://blowstack.com/tools/cloud-concepts-hub/synchronous-replication): Synchronous replication is a data-protection method that requires a write to satisfy replication acknowledgments from designated replicas before success is returned to the client.
- [Text Embeddings](https://blowstack.com/tools/cloud-concepts-hub/text-embeddings): Text embeddings are numerical vector representations of text that allow software to compare meaning and retrieve semantically related information.
- [Transaction Isolation](https://blowstack.com/tools/cloud-concepts-hub/transaction-isolation): Transaction isolation controls how concurrent database transactions observe and affect one another’s data.
