About replication in Cloud SQL

Replication is the ability to create copies of a Cloud SQL instance and offload work to the copies.

Introduction

The primary reason for using replication is to scale the use of data in a database without degrading performance.

Other reasons for replication include migrating data between regions.

Additionally, if an original instance is corrupted, a replica could be promoted to a standalone instance (in which case, existing replicas wouldn't consider that instance as primary).

When referring to a Cloud SQL instance, the instance that is replicated is called the primary instance and the copies are called read replicas. The primary instance and read replicas all reside in Cloud SQL.

When the first replica is created:

  • The primary instance is set to the full recovery model for all databases on the primary instance.
  • A temporary disk is created, and a full backup is taken and stored on the temporary disk. The temporary disk is deleted after the replica creation process is complete.

If, in the time period of the first replica creation, the user switches to the simple recovery model, then the replica creation fails.

The following apply to databases added to the primary instance after the replicas are created:

  • The databases are added to the availability groups automatically and are populated in the replicas using auto-seeding.
  • Each replica creation invokes a full (full recovery model) backup of databases on the primary instance. Logins and server objects created after the replica creation aren't replicated.

Cloud SQL supports the following types of replicas:

By using connector enforcement, you can enforce using only the Cloud SQL Auth Proxy or Cloud SQL Language Connectors to connect to Cloud SQL instances. With connector enforcement, Cloud SQL rejects direct connections to the database. You can't create read replicas for an instance that has connector enforcement enabled. Similarly, if an instance has connector enforcement enabled, then you can't create read replicas for the instance.

Cloud SQL doesn't support replication between two external servers.

Read replicas

You use a read replica to offload work from a Cloud SQL instance. The read replica is an exact copy of the primary instance. Data and other changes on the primary instance are updated in almost real time on the read replica.

Read replicas are read-only; you cannot write to them. The read replica processes queries, read requests, and analytics traffic, thus reducing the load on the primary instance. You can have up to 8 read replicas per primary instance.

You connect to a replica directly using its connection name and IP address. If you're connecting to a replica using a private IP address, you don't need to create an additional VPC private connection for the replica because the connection is inherited from the primary instance.

For information about how to create a read replica, see Creating read replicas. For information about managing a read replica, see Managing read replicas.

As a best practice, put read replicas in a different zone than the primary instance when you use HA on your primary instance. This practice ensures that read replicas continue to operate when the zone that contains the primary instance has an outage. See the Overview of high availability for more information.

Selecting an appropriate machine type

Read replicas can have a different machine type from that of the primary. You should monitor metrics on your instance such as CPU and memory usage to ensure that the replica instance is sized correctly for its workload, especially if it is smaller than the primary instance. A replica instance that is undersized is more prone to poor performance, such as frequent out-of-memory (OOM) events.

Cross-region read replicas

Cross-region replication lets you create a read replica in a different region from the primary instance. You create a cross-region read replica the same way as you create an in-region replica.

In the case of SQL Server read replicas, it is assumed that the replica is in the same virtual network as the primary replica, or that they communicate using Public IP.

Cross-region replicas:

  • Improve read performance by making replicas available closer to your application's region.
  • Provide additional disaster recovery capability to guard against a regional failure.
  • Let you migrate data from one region to another.

See Promoting replicas for regional migration or disaster recovery for more information about cross-region replicas.

Cascading read replicas

Cascading replication lets you create a read replica under another read replica in the same or a different region. Cascading replicas are implemented using distributed availability groups. Some use cases for cascading replicas are as follows:

  • Disaster recovery: you can use a cascading hierarchy of read replicas to simulate the topology of your primary instance and its read replicas. During an outage, your selected read replica is promoted to primary and the read replicas under the new primary continue to replicate and are ready for use. The old primary becomes a secondary of the new primary when it becomes available, and you can switch back to the old primary once it recovers using a switchover. For more information about using cascading replicas for disaster recovery, see About disaster recovery.
  • Performance improvement: Reduce the burden on the primary instance by offloading replication work to multiple read replicas.
  • Scale reads: You can have more replicas to share the read load.
  • Cost reduction: You can reduce networking costs by using a single cascading replica with cross-region replication in other regions.

Terminology

  • Cascadable replica: A cross-region read replica that can be used for switchover and replica failover operations in advanced disaster recovery (DR) with Cloud SQL for SQL Server.
  • Cascading replica: A read replica that can have its own replica.
  • Levels: You can create levels of replicas in a cascading replica hierarchy. For example, if you add four replicas to an instance, those four replicas are at the same level.
  • Sibling instances: Multiple replicas that replicate from the same primary instance. Siblings are at the same level in the replica hierarchy. A replica can officially have up to eight siblings.
  • Leaf replica: A read replica that does not have any replicas of its own. In a multi-level replication hierarchy, the leaf replica is the last level.
  • Promote An action that converts a replica, at any level in the hierarchy, into a primary instance. When promoted, the replica's cascading replica hierarchy is retained.

Configure cascading replicas

Cascading replicas let you add read replicas to any existing replicas. You can add up to four levels of replicas, including the primary instance. When you promote the replica at the top of a cascading replica hierarchy, it becomes a primary instance and its cascading replicas continue to replicate.

For more information on configuring distributed availability groups, see Configure an Always On distributed availability group.

Restrictions

  • You can't delete a replica that has replicas under it. To delete the replica, you must start with the leaf replicas and work your way upward through the hierarchy.
  • Circular region dependency isn't supported. To have the replica of a cascading replica in the same region as the primary instance, the cascading replica must also be in the same region.
  • You must create cascadable replicas in a different region than the region of the primary instance. Then you can create cascading replicas in the same region as the cascadable replica.

Replication use cases

The following use cases apply for each type of replication.

Name Primary Replica Benefits and use cases More information
Read replica Cloud SQL instance Cloud SQL instance
  • Additional read capacity
  • Analytics target
  • Improve read performance
  • Export from replicas
Cross-region read replica Cloud SQL instance Cloud SQL instance
  • Additional read capacity
  • Analytics target
  • Additional disaster recovery capability
  • Improve read performance
  • Migrate data between regions
SQL Server replication Instance external to Cloud SQL Cloud SQL for SQL Server instance
  • Migration path to Cloud SQL
  • Replication from an external server to Cloud SQL
  • Replication from Cloud SQL to an external server
  • Replication between Cloud SQL instances
  • Data replication to Google Cloud
  • Analytics target

Billing

  • A read replica is charged at the same rate as a standard Cloud SQL instance. There is no charge for the data replication.
  • Pricing for a cross-region read replica is the same as for creating a new Cloud SQL instance in the region. Refer to Cloud SQL instance pricing and select the appropriate region. In addition to the regular cost associated with the instance, a cross-region replica incurs cross-region data transfer charges for replication logs sent from the primary instance to the replica instance, as described in Network Egress Pricing.

Quick reference for Cloud SQL read replicas

Topic Discussion
Backups You cannot configure backups on the replica.
Cores and memory Read replicas can use a different number of cores and amount of memory from those of the primary instance.
Deleting the primary instance Before you can delete a primary instance, you must promote all of its read replicas to standalone instances or delete the read replicas.
Deleting the replica When you delete a replica, there is no impact on the status of the primary instance.
Deleting a replicated database You can delete a replicated SQL Server database using the Google Cloud console or the gcloud command, and the deletion automatically propagates to the replicas. You cannot delete a replicated SQL Server database using T-SQL commands.
Failover A primary instance can failover to a replica only if the replica is a DR replica. Read replicas are unable to failover in any way during an outage.
Load balancing Cloud SQL does not provide load balancing between replicas.
Maintenance windows Read replicas share maintenance windows with the primary instance. The replicas follow the maintenance settings for the primary instance, including the maintenance window, rescheduling, and the deny maintenance period. During maintenance, Cloud SQL updates all read replicas first before updating the primary instance.
Multiple read replicas You can create up to 8 read replicas for a single primary instance.
Private IP If you're connecting to a replica using a private IP address, you don't need to create an additional VPC private connection for the replica, as it is inherited from the primary instance.
Restoring the primary instance You cannot restore the primary instance of a replica while the replica exists. Before restoring an instance from a backup, or performing a point-in-time recovery on it, you must promote or delete all of its replicas.
Settings The settings of the primary instance are propagated to the replica, including changes to the data about users who can access the instance.
Stopping a replica You cannot stop a replica. You can restart or delete it, but you cannot stop it as you can a primary instance.
Upgrading a replica Read replicas can experience a disruptive upgrade at any time.
User tables You cannot make changes on the replica. All user changes must be done on the primary instance.

Limitations

  • This feature only applies to the following versions of Cloud SQL for SQL Server:

    • SQL Server 2017 Enterprise
    • SQL Server 2019 Enterprise
    • SQL Server 2022 Enterprise
  • Logins aren't propagated to a replica.

  • You must monitor your replica using T-SQL and/or SQL Server Management Studio.

  • Before deleting a database, you must close your database connections.

  • When you create a replica, the primary instance can contain no databases in single-user mode. Otherwise, replica creation fails.

What's next