When you purchase a Tair instance, you can choose between the cloud disk-based and local disk-based deployment modes based on your business requirements. This topic compares Tair instances that use local disks and those that use cloud disks.
Overview
Item | Cloud disk-based instance (recommended) | Local disk-based instance |
Architecture | Uses the new-generation management architecture that offers greater flexibility and enhanced scalability. This architecture serves as a basis for subsequent development. | Uses the conventional management architecture. Cluster instances come with predefined specifications that cannot be changed. |
Scalability |
|
|
Feature comparison
Cloud disk-based instances are built on cloud-native architecture. Cloud disk-based instances that use the cluster architecture support imperceptible scaling. Tair embraces cloud-native architecture as the primary basis for subsequent development.
In the following table, ️✔️ indicates that this feature is supported, ❌ indicates that this feature is not supported, and ➖ indicates that this feature is not applicable.
Differences
Category | Feature | Classic (local disk-based) instance | Cloud-native (cloud disk-based) instance | ||||
Standard architecture | Cluster architecture | Read/write splitting architecture | Standard architecture | Cluster architecture | Read/write splitting architecture | ||
Lifecycle management | ➖ | ❌ | ➖ | ➖ | ✔️ | ➖ | |
➖ | ➖ | ❌ | ➖ | ➖ | ✔️ | ||
Bandwidth management | ✔️ | ✔️ | ✔️ | ❌ | ❌ | ❌ | |
Extended features | ✔️ | ✔️ | ✔️ | ❌ | ❌ | ❌ |
All features
Category | Feature | Local disk-based instance | Cloud disk-based instance | ||||
Standard architecture | Cluster architecture | Read/write splitting architecture | Standard architecture | Cluster architecture | Read/write splitting architecture | ||
Lifecycle management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ Supports imperceptible scaling. | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
➖ | ❌ | ➖ | ➖ | ✔️ | ➖ | ||
➖ | ➖ | ❌ | ➖ | ➖ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Unsubscribe from subscription instances | ❌ | ❌ | ❌ | ❌ | ❌ | ❌ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Network connection management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
➖ | ✔️ | ➖ | ➖ | ✔️ | ➖ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Bandwidth management | ✔️ | ✔️ | ✔️ | ❌ | ❌ | ❌ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
High availability management | Manually switch workloads from a master node to a replica node | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ |
➖ | ✔️ | ✔️ | ➖ | ✔️ | ✔️ | ||
➖ | ✔️ | ✔️ | ➖ | ✔️ | ✔️ | ||
Parameter management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
Tag management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
Management of maintenance, migration, and data export | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Security management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Performance monitoring | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Log management | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Backup and restoration | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | |
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ️️️️️️✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
Extended features | ✔️ | ✔️ | ✔️ | ❌ | ❌ | ❌ | |
➖ | ✔️ | ✔️ | ➖ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ||
✔️ | ✔️ | ✔️ | ✔️ | ✔️ | ✔️ |