PostgreSQL Replication Services

Protect your database with state-of-the-art PostgreSQL Replication.

Modern replication is almost a standard now—don’t be left behind in case of disaster, and ensure the security of your database investment. Implement an expert-level PostgreSQL replication solution that only XTIVIA can provide.

What is PostgreSQL Replication?

Replication is the duplication of data from one place to another. Data is transported by programs to another location, and then loaded. A more efficient alternative to basic replication is to extract only the changes since the last processing cycle and transport/apply those to the receiving location. In most cases, replication must not interfere with existing applications and have minimal impact on production systems. Thus, replication processes must be managed and monitored.

What We Provide

XTIVIA’s PostgreSQL DBAs start by connecting with your business to develop a PostgreSQL replication strategy. Whatever your organizational objectives, our staff comes equipped to confidently handle a full spectrum of replication server activities.

These functions are designed to help you implement and maintain a solution that works best for your company.

Our services include:

Initial assessments and requirements review
Replication system design
Approach and plan definition
Schedule and cost estimations
Installation
Testing and validation
Training to administer replication server
Replication server monitoring using Virtual-DBA when requested

Assess your PostgreSQL replication

We provide value to your business with assessment and assistance for your PostgreSQL database and DBA team.

Start Your Review Today!

Types of Replication

Shared Disk Failover

This method allows for rapid failover without data loss, while still having only one copy of the database. This method avoids any synchronization overhead by sharing a common disk between all database servers. If the main server fails, the standby server accesses the shared disk as though it was recovering from a database crash. The main point of failure is the shared disk array; if it becomes corrupt or unavailable, primary and standby are non functional. Also, only the primary server should have access to the storage while it is active.

File System Replication

In this type of replication, hardware is shared, and all changes to the file system need to be mirrored on another server. The main challenge here is that primary file system mirroring is needed to guarantee the standby has a consistent copy of the file system. Writes to the file system need to be done in exactly the same order on the standby as they were on the primary.

Hot/Warm Standby Using “Point In Time Recovery” (PITR)

Logs from the primary server are applied to the warm standby server. The standby can be made the new primary and would contain all the changes from the logs transferred before the failure. This method is asynchronous and can only be performed against the complete primary server.

Master-Slave Replication

All queries that modify data are sent from the master asynchronously to the slave. In this setup, the slave is available for read only queries while the master is running.

Statement-Based Replication Middleware

Each server operates independently, while a middleware program captures each SQL query and forwards it to all or just one of the servers. Read-only queries can be sent to just a single server, allowing for a distributed workload. Read/write queries are sent to all servers. Extreme caution should be used to make sure all transactions commit or abort on all servers. Also any application or middleware should be written in a manner so as to not use unmodified functions or sequences that could potentially return different values depending on which server was used.

Asynchronous Multimaster Replication

Each server works on its own, communicating with the other servers to identify conflicting transactions. Rules are defined or users can resolve any conflicts. This works well in an environment where servers maintaining a constant connection is a challenge.



PostgreSQL Remote Database Services

l

PostgreSQL Health Check and Performance Tuning

w

What's Next?

We would love to help you with your database project or manage your environment, call us 888.685.3101 Ext. 2 or complete the form and start a conversation with Virtual-DBA today!