Overview
You can create backups for GeminiDB Redis instances to ensure data reliability.
Usage Notes
- Backup data cannot be downloaded or exported.
- GeminiDB Redis API supports only full backup.
- You can back up data of instances that are in the Available state.
- During data backup, only a small portion of CPUs is used. Uploading backups to OBS will consume compute node bandwidth, which may cause slight jitter in latency.
- After an instance is deleted, the manual backups are retained until you manually delete them.
- After an instance is deleted:
- If the instance has automated full backups, the system keeps the most recent automated full backup of the previous day. If there is no automated full backup generated on that day, it retains the latest available one and moves it to the recycle bin.
- If there are fewer than 100 backups in the recycle bin, the system will create a full backup and move it to the recycle bin.
- If any backups in the recycle bin exceed the retention period set in the recycling policy, the system automatically deletes them. If you delete an instance, its automated backups that are not in the recycle bin will also be deleted.
Backup Methods
GeminiDB Redis instances support both automated and manual backups.
| Method | Description |
|---|---|
| You can modify the automated backup policy on the GeminiDB console. The system automatically backs up instance data based on the configured time window and backup cycle, and retains the backups for the specified duration. Automated backups cannot be manually deleted. You can adjust their retention period by following the instructions provided in Modifying an Automated Backup Policy. Expired backups will be automatically deleted. | |
| Manual backups are user-initiated full backups of DB instances and can be retained until you manually delete them. You can trigger a manual backup at any time to meet your service requirements. You are advised to back up your DB instance regularly. If the DB instance fails or data is corrupted, you can restore data from backups. |
How Data Is Backed Up
GeminiDB Redis API uses an architecture with decoupled storage and compute. Figure 1 shows how GeminiDB Redis API backs up data. GeminiDB Redis API takes snapshots of data in the storage pool in seconds and transmits them to the compute layer, which will then pass them to OBS. Snapshots are stored as compressed files in OBS buckets and do not occupy any storage space of your instance. Snapshot creation and deletion have no impacts on the computing layer. When snapshots are uploaded, some compute resources are consumed, so the instance CPU and memory usage may go high.
GeminiDB Redis API backs up data faster than Redis and does not produce jitter.
Backup Storage
Backups are stored in OBS buckets to provide disaster recovery and save storage space.
After you purchase an instance, GeminiDB Redis API provides additional backup storage equal in size to the purchased storage, free of charge. For example, if you purchase an instance with 100 GB of storage, you will get an additional 100 GB of backup storage at no extra cost. If backup data stays within 100 GB, it is stored in OBS free of charge. If it exceeds 100 GB, you will be billed for the additional space at standard OBS rates. For details, see OBS Pricing Details.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot
