LindormTSDB is a time series database optimized for high-throughput workloads—reading millions of data points per second and writing tens of millions per second. Core capabilities include:
Four write protocols: SQL INSERT (JDBC), InfluxDB line protocol, OpenTSDB API, and Prometheus remote write
Compression to 1–2 bytes per data point, reducing storage usage by 90%
Cold data archiving that cuts storage costs by more than 70%
SQL-based downsampling, interpolation, and spatial aggregation
In-database machine learning (Lindorm ML) via standard SQL—no ML expertise required
Network isolation, IP whitelisting, and database-level permission control
Efficient reading and writing for time series data
Data writing
LindormTSDB accepts data over four protocols:
SQL INSERT statement — Connect via Java Database Connectivity (JDBC). This is the recommended method.
InfluxDB line protocol — Drop-in compatibility for existing InfluxDB write pipelines.
OpenTSDB API — Compatible with OpenTSDB clients and tooling.
Prometheus remote storage — Send monitoring data from Prometheus using the remote write API. LindormTSDB acts as a remote storage backend.
Data query
Execute SQL statements or use the OpenTSDB API to query data. The Lindorm console also provides a built-in data query interface for visualizing results with data groups, downsampling, and spatial aggregation.
Efficient time series data storage technologies
Time series data compression
LindormTSDB compresses time series data to an average of 1–2 bytes per data point, reducing overall storage usage by 90% while also increasing write throughput.
Cold data archiving
Archive cold data to cost-effective long-term storage, cutting storage costs by more than 70%.
Data management
LindormTSDB gives you SQL-level control over how data is stored and who can access it.
Database configuration — Use SQL to set partition intervals, cold/hot data boundaries, and time-to-live (TTL) values. Data in different databases is isolated from each other.
User and permission management — Manage users with identity verification and permission checking. Configure access via SQL or the Lindorm console.
For database configuration details, see Database management. For user and permission configuration, see Manage users and permissions.
Computing capabilities for time series data
The Lindorm console supports advanced time series queries, including:
Downsampling — Aggregate high-resolution data into lower-resolution summaries.
Data interpolation — Fill gaps in sparse or irregular time series.
Spatial aggregation — Aggregate data across spatial dimensions (a LindormTSDB-specific capability).
Monitoring and O&M
View instance status, performance metrics, and storage usage in real time. Configure alert rules and notification methods to catch resource bottlenecks early.
Data security and cluster security
LindormTSDB protects data at every layer—network access, connection filtering, and permission control work together to prevent unauthorized access:
Network isolation — Access instances over virtual private clouds (VPCs) to prevent unauthorized network exposure.
IP whitelisting — Add allowed client IP addresses to an instance whitelist. Clients not on the whitelist cannot connect, even from within the same VPC.
Permission control — Manage user permissions at the database level. Use whitelists together with permission control for fine-grained access management.
Data availability — Multiple replicas are created by default to ensure data availability.
In-database machine learning
LindormTSDB provides an out-of-the-box in-database machine learning service called Lindorm ML. Run the full machine learning workflow using standard SQL statements—no specialized ML knowledge required.
Key advantages:
No data export — Models and data stay in the database, so there is no need to move data to external platforms. This speeds up iteration and helps meet regulation requirements.
Enterprise-grade controls — Because models and data reside in the database, you get the same permission management, data audit, and data encryption that apply to all your database data.
Lindorm ML supports common time series forecasting and anomaly detection algorithms.