How to Achieve Success During an Oracle to MariaDB Migration

MariaDB is a powerful open-source database technology that offers a range of enterprise-grade features and benefits, making it an attractive migration option for organizations using Oracle. However, you need to develop a comprehensive Oracle migration plan for a successful move. Here are some lessons Datavail has learned over thousands of database migrations.

Assess Your Cost Savings

Migrating from Oracle to MariaDB is a popular way for organizations to reduce their costs because of high expenses associated with licensing, add-ons, and long lead times on feature requests. According to MariaDB, “the total cost of Oracle is 84x higher than the MariaDB Platform, and organizations can save over $9 million after three years by choosing the MariaDB Platform.”

MariaDB offers an open-source model that lowers support costs and speeds up the process for feature requests. If you have the technical resources, you can develop custom database features and add them to MariaDB yourself. You also keep a comparable level of functionality, stability, and professional support when you migrate.

Perform Performance and Functional Testing at Scale

A successful and thorough proof of concept is an essential part of an Oracle to MariaDB migration. This proof of concept forms the basis of the migration plan and allows you to identify any roadblocks that could lead to failure. To get the most out of your testing, you should:

• Use the same hardware as your production environment.
• Emulate your application and database environment as much as possible.
• Test against a product size data set.

Choose the Right Hardware Specifications

Trying to run MariaDB databases on non-database optimized hardware or those smaller than your Oracle environment can cause a performance bottleneck. When you select your MariaDB hardware, ensure that the following components have the right capabilities for your database load and application usage:

• Types of drives
• IOPS capacity
• Drive mount options
• RAID or other drive redundancy options
• Memory size and type
• CPU capacity

Leverage Oracle SQL Mode in MariaDB

One of the most significant advantages that MariaDB brings to the table over other open-source database options is its built-in Oracle SQL Mode. This feature helps companies quickly migrate from Oracle to MariaDB without completely changing their application code. Oracle Mode supports almost all PL/SQL syntax and commands. Most times, you end up making minimal code changes to make an application MariaDB compatible.

Understand MariaDB’s High Availability Architecture Gains

MariaDB’s overwhelmingly lower cost opens up more options for High Availability (HA) architecture. During one of our recent migration projects, our customer took a three-node DataGuard Architecture between two data centers and doubled their HA footprint with standard MariaDB Replication across six database servers.

Previously, this customer only had two nodes within the primary data center region. Any outages or planned maintenance left them vulnerable to outages or forced failover to the secondary datacenter if something happened to the remaining standing node.

The extra nodes added a layer of HA that helped the customer’s peace of mind and strengthened the availability and stability of their database architecture. This infrastructure also provides more scalability with MariaDB’s Maxscale Read-Write splitting between the nodes. For comparable architectures to Oracle RAC, MariaDB Galera Cluster provides multi-master clustering, which can combine with Replication to the disaster recovery site.

Adding Load Balancing Through MariaDB MaxScale

The MariaDB suite of products, included with the Enterprise license, offers the MaxScale advanced proxy server between the application and the database servers. This tool is full of useful and impressive features that make it an essential part of any MariaDB architecture, ranging from data masking to basic connection routing, automating database failover and replica promotion. MariaDB MaxScale 2.5 also introduces the MaxScale GUI, which gives you a graphic user interface for a more user-friendly experience compared to the command-line utility or working directly in the configuration file.

You can configure MaxScale with Query Caching using Native Caching, Redis, or MemCached, which can improve repetitive query speed by 8x and lighten the load on your database.