Hot and cold data separation reduces storage costs by moving infrequently accessed data to lower-cost storage while keeping recent, actively accessed data on high-performance storage. Lindorm supports this capability across multiple engines. Select the engine you use and follow the linked guide.
| Engine | Approach | When to use | Documentation |
|---|---|---|---|
| LindormTable | Separate hot and cold data based on a custom time column | Your table has a business timestamp column (for example, event_time) that accurately reflects data recency | Separately store hot data and cold data based on custom time columns |
| LindormTable | Separate hot and cold data based on the built-in row timestamp | You rely on the system-assigned write timestamp rather than a business-defined time field | Separately store hot data and cold data based on timestamps |
| LindormTSDB | Archive cold time series data to lower-cost storage | You store time series data with predictable aging patterns and want automatic archiving | Cold data archiving |
| LindormSearch | Move infrequently accessed search index data to cold storage | You have search indexes that are rarely queried and want to reduce index storage costs | Configure cold storage |
| Underlying file | Separate hot and cold data at the file level | — | Documentation not yet available |
该文章对您有帮助吗？