A crucial topic you’ll need to understand is database replication, specifically the use of read replicas. In this post, we’re going to dive deeper into this subject, discussing what database replication is, why it’s beneficial, and providing examples with AWS RDS service as a practical implementation.

Database Replication & Its Significance

Database replication is the method of copying data from a database in one server (primary database) to a database in another server (replica), with the intent of increasing data availability and creating a single-user environment where users can work with the replicated data without impacting others.

Database replication offers several benefits:

  • Improved Performance: Read replicas can handle read-only traffic, thereby reducing the load on the primary database and improving overall performance.
  • Data Protection: Replicas act as backups in case the primary database fails, ensuring data persistence.
  • Geographical Distribution: Replicas can be geographically distributed to serve users from the nearest location, reducing latency.

Understanding Read Replicas

In the context of AWS services, a read replica refers to a standalone DB instance that is a copy of the main database. Data is asynchronously copied from the primary database to the read replica, which can then handle user demand for read-heavy database workloads.

Read Replicas in AWS RDS

Amazon Relational Database Service (RDS) is a web service that simplifies setting up, operating, and scaling a relational database in the cloud. AWS RDS supports creating read replicas, which can enhance scalability and availability of read traffic.

Let’s look at a simple example of how to create a read replica of an RDS DB Instance in AWS, using the AWS Management Console.

First, navigate to the “RDS Dashboard”, then select “Databases” and choose the primary DB Instance for which you want to create a read replica.

Next, choose the “Actions” drop-down menu, select “Create read replica”, and fill in the necessary information including replica name, instance size, storage type etc.

Finally, choose “Create read replica”, and within a short period of time (depending on the data), your read replica will be available and ready for read traffic.

AWS RDS supports read replicas for MySQL, MariaDB, PostgreSQL, Oracle, and SQL Server.

Monitoring Replication Lag

One caveat with replication, in general, is the potential for “Replication Lag.” This is the time lag between a write happening on the primary database and the same write appearing on the replica. It’s important to monitor this lag to ensure that your application serving data from the replica is not serving stale data.

AWS CloudWatch provides a metric named ‘ReplicaLag’ which can help to monitor this. The ‘ReplicaLag’ metric is the difference in seconds between the source DB instance write timestamp (recorded in the transaction logs) and the read replica write timestamp (replayed from the transaction logs).


Understanding database replication and the use of read replicas is a vital part of preparing for the AWS Certified Solutions Architect – Associate examination. As data demands increase, robust and efficient strategies such as database replication are essential for ensuring that data is accessible, secured, and served up quickly.

AWS services like RDS offer handy built-in features that make creating and managing read replicas relatively easy. These features, combined with monitoring services like CloudWatch, make AWS a useful platform for implementing replication strategies.

Practice Test

True or False: Database replication increases the redundancy of information and improves the reliability and accessibility of data.

Answer: True

Explanation: Database replication involves making copies of a database to ensure its availability even in cases of data loss, hardware failure or other issues.

What is a read replica in AWS RDS?

A. A copy of the main database for writing operations
B. A secondary backup of the main database
C. A copy of the main database primarily for read traffic
D. A clone of the application for load balancing

Answer: C. A copy of the main database primarily for read traffic

Explanation: A read replica in AWS RDS is intended for offloading read traffic from the main DB instance. It does not support write operations.

True or False: You can create a read replica of a read replica in AWS?

Answer: True

Explanation: AWS allows the creation of second-tier read replicas, i.e., a read replica of a read replica. This can be used for various reasons such as offloading your read traffic even further.

Which AWS database service does not support read replicas?

B. DynamoDB
C. DocumentDB
D. Redshift

Answer: B. DynamoDB

Explanation: By default, DynamoDB does not support read replicas. It manages database replication on its own for its provisioned and on-demand capacity modes.

True or False: AWS RDS supports cross-region replication.

Answer: True

Explanation: AWS RDS does support the creation of read replicas in a different region than that of the source database. This is especially useful in disaster recovery scenarios.

Replication in databases can improve:

A. Data availability
B. Data durability
C. Performance
D. All of the above

Answer: D. All of the above

Explanation: Replication increases data availability by keeping multiple copies of data. It enhances data durability by preventing data loss, and it improves read performance by allowing reading from multiple copies.

True or False: A read replica supports both read and write operations.

Answer: False

Explanation: A read replica is primarily for handling read traffic. It does not support write operations.

When a read replica is promoted to a standalone DB instance, it __________.

A. Retains the data of the source DB instance
B. Deletes all data
C. Copies all data from another read replica
D. None of the above

Answer: A. Retains the data of the source DB instance

Explanation: When a read replica is promoted, it keeps the data of the source DB instance up to the time of promotion completion.

True or False: Read replicas can only be used with a Multi-AZ DB instance.

Answer: False

Explanation: Read replicas in Amazon RDS can be used with both single-AZ and Multi-AZ DB instances.

In AWS RDS, when you delete a source DB instance with read replicas, ________.

A. The read replica is promoted to a standalone DB instance
B. The read replica is deleted
C. The read replica is orphaned
D. The read replica becomes the source DB instance

Answer: C. The read replica is orphaned

Explanation: If you delete a source DB instance, its read replicas remain intact and become orphaned. They are not automatically promoted or deleted.

Interview Questions

What is database replication in AWS?

In AWS, database replications the process of storing and maintaining multiple copies of the same data in separate databases in different regions to ensure high availability, durability and data integrity.

What is Read Replica in AWS RDS?

A Read Replica is an additional instance of a database in Amazon RDS which serves as a standby copy of the primary database. It can be used for read-heavy database workloads to offload the burden from the primary database instance.

How can read replicas improve the performance of your database in AWS?

Read replicas can enhance database performance by distributing the read traffic among multiple databases rather than a single primary database. This allows for higher scalability and availability of applications by separating the read and write workloads.

Are read replicas available in all AWS RDS database engines?

No, Read replicas are currently supported in RDS for MySQL, MariaDB, PostgreSQL, and Aurora.

Can you create read replicas of read replicas in AWS RDS?

Yes, it’s possible to create read replicas of existing read replicas in AWS RDS. However, latency can be increased because all write operations have to be replicated through the primary replica.

Are Read replicas in RDS read-only or can they be written to as well?

Read replicas in AWS RDS are read-only copies of the primary database. They are not designed to take write operations.

How many read replicas can you create in AWS RDS?

You can create up to five read replicas for each primary DB instance in AWS RDS.

Can you make your Read Replica in AWS RDS as your primary database?

Yes, in the event of a primary DB instance failure, you can promote a Read Replica to act as your primary DB instance.

What happens to the data in a read replica when it’s promoted to a primary database?

When a read replica is promoted to a primary database, it is detached from the source database, and all data in the read replica becomes independently writable.

Does database replication impact the performance of the primary DB instance in AWS RDS?

No, the performance impact on the primary DB instance is minimal as the data replication is done in the storage layer of the DB instance, which is transparent to the DB instance itself.

How does AWS handle replication lag in read replicas?

AWS uses MySQL and PostgreSQL’s built-in replication functionality to create replicas, hence, any lag between the source DB Instance and read replica is typically minimal.

What happens if a primary DB instance fails in AWS RDS with read replicas?

When the primary DB instance fails, one of the read replicas can be promoted to become the new primary DB instance, thus minimizing the downtime of your application.

Can you use read replicas in AWS RDS to enhance database backup?

Yes, you can offload your backup processes to a read replica so that your primary DB instance is not affected during backup operations.

What is Multi-AZ deployment in AWS RDS and how is it different from Read Replicas?

Multi-AZ deployment in AWS RDS is a replication method that automatically creates and maintains a synchronous standby copy of the primary DB in a different Availability Zone. Unlike Read Replicas, which are primarily used to improve read performance, Multi-AZ deployment is designed to provide high availability and failover support.

Can Read Replicas be created in a different region than the primary DB in AWS RDS?

Yes, Read Replicas can be created in a different region than the primary DB instance, offering a way to scale read traffic or to serve global users.

Leave a Reply

Your email address will not be published. Required fields are marked *