A placement group is a policy for distributing instances on the underlying hardware. The instances you create in the placement group feature disaster recovery and high availability when launched. Tencent Cloud CVM provides an instance placement policy, which can force the instances to be dispersed with a certain policy and reduce the impact of the underlying hardware/software failure on CVM services. You can use the placement group to deploy CVM instances on different physical servers to ensure service high availability and the underlying disaster recovery capability. When you create instances in a placement group, we will launch them in a specified region according to the deployment policy you configured in advance. If you did not configure a placement group for the instances, we will try to launch them on different physical servers to ensure service availability.
Currently, spread placement group is supported. A spread placement group is a group of instances that are placed on different underlying hardwares and have high availability. We recommend that you use a spread placement group for applications of important instances that need to be placed separately, such as master/slave databases and high-availability clusters. By launching instances in a spread placement group, you can minimize the simultaneous failure of instances with the same underlying hardware.
A spread placement group is region specific and can be deployed across multiple availability zones. There is a limit on the number of instances that can be placed in a group. For more information, please see the Console.
When you launch instances in a spread placement group, the request will fail if you have insufficient hardware. You can wait for a while and try again.
Before using a spread placement group, note the following rules:
For more information on operations, please see Spread Placement Group.