Managed Database

Arvancloud Database is a managed service for creating database clusters. This product helps you build a cluster of configured databases in just a few minutes without having to go through the process of installing, configuring, maintaining, and securing databases manually.

Arvancloud database clusters can include Standby and Read-only nodes for high availability and performance and can be scaled quickly. Also, full backups of your data are taken daily and automatically with PITR. PITR allows you to roll back database data to a specific point in the past. Therefore, even databases with one node are equipped with automatic failover.

Database Engine

Currently, it is possible to build database clusters with MySQL engine, and in the future, building clusters with MongoDB, PostgreSQL, and Redis will be available.

MySQL is one of the most common and popular open-source and relational database management systems. This engine can provide high speed and reliability for saving and retrieving data.

Cluster Configuration

Arvancloud database clusters can include the following nodes:

• Primary node

  This node is responsible for processing queries and performing CRUD operations on the database and is the main source of data storage and retrieval in the cluster. So, standby and read-only nodes sync data with this node.

• Standby nodes

  Standby nodes store a copy of the main node's data and are always in sync with it to replace it if there is a problem with the primary node. This replacement happens in the fastest possible time and with the least downtime. When creating a new cluster, you can add one or two standby nodes.

• Read-only nodes

  These nodes are a copy of the primary node, but they only respond to read requests and cannot make changes to the databases. They are used for horizontal scaling of the database and you can use this node in applications where the request to read data is high. For each primary node, you can add up to 3 read-only nodes.

A database cluster consists of a primary node and standby nodes. Read-only nodes are not part of the cluster; This means that you can access it independently from a different link or change the size of its resources.

All Arvancloud database clusters have an automatic failover mechanism and when an error occurs in the primary node, it will be replaced with another node. However, to achieve High Availability and reduce service recovery time, it is highly recommended to add one or two standby nodes to your cluster.

If you do not add a standby node to the cluster and the primary node fails, the cluster will be unavailable until the replaced node is activated. This time can vary based on the amount of data in your cluster, and the larger the data, the longer the cluster recovery time.

To achieve maximum availability, add two standby nodes to the cluster so that if the primary node and one of the standby nodes fail, the service doesn't become unavailable.

Transferring traffic to a standby node takes less than a minute, but replacing a primary node is time-consuming. In fact, the availability of your database and data depends on your configuration.