1. Home
  2. Docs
  3. CCX
  4. How-To
  5. Deployment
  6. Deploying a database in AWS

Deploying a database in AWS

CCX allows you to deploy the following types of database clustering systems:

  • PostgreSQL + HAProxy
  • MariaDB Cluster + ProxySQL
  • Percona XtraDB Cluster + ProxySQL

Database

Configure the database name, version and topology.

Field Description
Data Store Name
  • The name of this data store. This name can NOT be changed later.
Tags
  • Specify tags to allow filtering and searching for clusters. Each cluster can have zero or many tags to help keep the clusters organized. Note that special characters like spaces, tabs and dollar signs are not supported.
  • The created tags can be filtered while looking up the clusters in the CCX dashboard page, by clicking on the magnifier glass icon on the top menu.
Select Database Vendor
  • Supported database versions and vendor:
    • MariaDB Cluster f10.5 by MariaDB
    • Percona XtraDB Cluster 5.7 by Percona
    • PostgreSQL 12 by PostgreSQL
Select Number of Nodes
  • The number of nodes indicates the topology that CCX will deploy:
    • 1 node – Useful for development purposes. No high availability.
    • 3 nodes – Recommended for the production usage with high availability features.
  • A database load balancer will be co-located on every database node, including the 1-node setup.

Cloud

Configure the cloud provider and region.

Field Description
Select Cloud Provider
  • Choose the supported cloud providers.
Select Region
  • Choose a region where you would want the database cluster to be deployed.

Instance

Configure the cloud instance size and volume.

Field Description
Select Instance Size
  • Choose the supported instance size offered by CCX. The instance sizes are independent of the cloud provider’s.
Volume Type
  • Choose the supported volume type for the selected cloud platform.
    • GP2: General Purpose SSD
    • IO1: Provisioned IOPS SSD (99.8% – 99.9% durability)
Volume Size (Gib)
  • The volume size. The minimum and maximum database sizes are depending on the volume type.
    • GP2 – 1 GiB – 16 TiB
    • IO1 – 4 GiB – 16 TiB
IOPS
  • The input/output operations per second (IOPS). The maximum values are depending on the volume type.
    • GP2 – 16,000
    • IO1 – 64,000
  • If your production workload requires high OLTP, and fast, consistently high throughput performance, you should configure your DB instance with IO1 or IO2 volumes.

Network

Configure the network and availability zones. Note that network settings cannot be modified in the future.

Field Description
Network
  • Choose the networking type:
    • Private: All nodes will be configured with a private IP address. To connect to the data store, you need to peer with the created VPC. See VPC and VPC Peering.
    • Public: All instances will be deployed with public IP addresses. Access to the public IP addresses is controlled by a firewall.
Network Type
  • Choose the high availability zone:
    • High Availability – Partition the nodes on now ore more availability zone in the selected region.
    • Regular – Deploy all nodes in one availability zone.
Select VPC
  • Choose the VPC of that particular region. Only available if you choose Private under the Network.
  • If you do not see the VPC in the drop-down, create a VPC beforehand. See VPC and VPC Peering.
Select Availability Zones
  • Choose the availability zone (AZ) for every node in the cluster. Only available if you choose High Availability under the Network Type.
  • For high availability, choose a different zone for every node. If you plan to run all nodes in one availability zone, choose Regular under the Network Type.

Overview

Summary of the deployment.

Field Description
Create Data Store
  • Starts the database deployment. You will be forwarded to the CCX dashboard where you can monitor the cluster status.
  • The following are the possible status:
    • Deploying – The deployment is in progress.
    • Failed – The deployment fails. Our technical support team will be notified of any deployment failures. You may delete the failed deployment, or create another deployment. A failed cluster will not be charged.
    • Available – The deployment has successfully completed and the cluster is running.
    • Unreachable – The deployment succeeds, but the cluster is not running.
Was this article helpful to you? Yes No