DocsICP FilingConsole
官方文档
首页

Introduction

更新时间：
复制 MD 格式
产品详情
我的收藏

LindormTSDB collects, stores, queries, and analyzes time series data at scale. It is designed for workloads that require high-throughput writes and low-latency queries, with built-in compression, multi-dimensional indexing, and in-database machine learning.

Core capabilities

CapabilityDetails
High performanceIn some scenarios, a single core handles ≥100,000 data points per second on write, ≥20 million queries per second, and >1,000 batch write and update operations per second.
Data compressionSelf-developed time series compression combined with general block compression achieves up to a 10:1 compression ratio.
Time series indexingMulti-dimensional data queries across tens of billions of time series.
Time series computing10+ aggregation functions, 20+ fill policies, and 10+ interpolation algorithms.
Auto ScalingDistributed architecture with online elastic scaling to handle any data volume.
In-database machine learningBuilt-in time series forecasting and anomaly detection.

For the full feature list, see Features.

Use cases

LindormTSDB is suited for high-throughput, low-latency workloads across large time series datasets:

  • Monitor IoT device telemetry: ingest tens of millions of data points in seconds from sensors and devices, with high compression ratio and low-cost storage, pre-downsampling, and visualized query results for anomaly detection in real time

  • Analyze IIoT (Industrial Internet of Things) equipment data: store and query machine metrics to streamline equipment management and predictive maintenance

  • Track application performance: collect and aggregate APM (application performance monitoring) metrics with multi-dimensional aggregate computing and interpolation to identify bottlenecks and improve system reliability

Connect to LindormTSDB

LindormTSDB supports SQL-based access. Choose the method that matches your role:

Previous：LindormTSDBNext：Features