English

Asymmetric Deployment

Asymmetric deployment is designed for enterprise users with resource constraints, high availability requirements, or specific business scenarios.

Before deploying the database, refer to the cluster deployment procedures for your operating system to complete prerequisites such as installation preparation and database package installation.
For CentOS 7 and RedHat 7, click here.
For CentOS 8 and RedHat 8, click here.
For Ubuntu 20.04, click here.

1 Database Deployment

1.1 Log in to the UI

When no cluster is deployed, access the graphical installation wizard via a browser using the following URL. The server whose IP address you enter will become the Master node of your cluster (in this example, mdw):

http://<IP>:8240/

On the first page of the installation wizard, provide the superuser password. You can view it using the sudo more /etc/matrixdb6/auth.conf command.

deployer_1

1.2 Select Multi-Node Deployment

deployer_2

1.3 Add Nodes

In this example, we add one node: sdw1.

deployer_3

Enter the IP address, hostname, or FQDN of sdw1 in the text box and click "OK".

deployer_4

deployer_5

1.4 Configure Cluster Parameters

Configure cluster parameters. In this example, set sdw1 as the standby node, enable data mirroring, and set the number of instances per node to 2 primary + 2 mirror. These settings will adjust the cluster according to balanced deployment mode (standard cluster configuration).

deployer_6

1.5 Set Storage Paths

Set the data storage path and etcd path.

deployer_7

1.6 Confirm Execution Information

This section describes the key steps for asymmetric deployment.

The deployment execution page consists of three tabs: Basic Information, Node Information, and Instance Distribution.

1.6.1 Basic Information

The Basic Information tab displays connection details, cluster size, and high availability settings for the current deployment plan.

deployer_8

1.6.2 Node Information

The Node Information tab lists each node's role and directory information.

deployer_9

1.6.3 Instance Distribution

The Instance Distribution tab shows the topology of all instances in the data cluster. Instances include Master, Standby, Primary, and Mirror. Large blocks represent nodes; small numbered blocks represent instances, where the number indicates the content ID.

deployer_10

Hover over an instance block to view its details.

deployer_11

Click "Advanced Operations", then "Customize Instance Distribution" to view the raw configuration file.

deployer_12

The instance distribution configuration file is in CSV format. Each line represents one instance configuration with four columns: content ID, hostname, port number, and data directory.

deployer_13

1.6.4 Rules for Instance Distribution Configuration File

Customizing the instance distribution file must strictly follow these rules.

You must manually edit the distribution rule file by modifying the hostname, port, and data directory associated with each content ID to define your desired asymmetric cluster topology.

The first column is the content ID. It must be a sequence of integers starting from -1.
- -1 may appear once or twice. The first occurrence represents the master; the second represents the standby. More than two occurrences will result in an error.
- Content IDs 0 and above may appear once or twice. However, the count of each content ID must be consistent across the cluster. For example, if content ID 0 appears twice, then content ID 1 must also appear twice. The first occurrence of any content ID ≥ 0 represents a primary segment; the second represents its mirror.
The second column is the hostname. It must be within the list of hostnames provided in the template. You cannot add new hostnames directly here. To add a new host, use the earlier step [1.3 Add Nodes].

Note: Do not change the hostname of the first row (master), as the current browser session and backend deployment commands are running on this machine. Changing it may cause unexpected errors.
The third column is the port number. Ports must be unique on the same host.
The fourth column is the data directory. This refers to the disk mount point, not the final data directory path. For example, entering only /data2 in the configuration file results in the actual data directory being /data2/mxdata_xxxxxxxxxxxx/master/mxseg-1/ after deployment.

Note: Modifying the fourth column affects only the corresponding segment’s data directory. The etcd storage path still follows the selection made in [1.5 Set Storage Paths].

After editing the configuration file, click "OK". Upon successful save, the updated instance distribution will be reflected in the topology diagram.

deployer_14

deployer_15

1.7 Execute Deployment

Click "Deploy" to start cluster deployment and view progress.

deployer_16

1.8 Deployment Success

After successful deployment, a Access Guide page appears, describing how to connect to the database.

deployer_17

Asymmetric deployment is now complete. Click "Finish" to go to the UI homepage.

Others

Supports deployment with blank machines (etcd-only nodes)
In OLTP deployments, a three-node topology can be achieved:
```
  mdw: master, seg0 (no workload), etcd0
  smdw: standby, etcd1
  sdw1 (low-spec hardware): etcd2
```

← Previous

Upgrade - Minor version

Kubernetes