// the find
taosdata/TDengine
High-performance, scalable time-series database designed for Industrial IoT (IIoT) scenarios
TDengine is a time-series database written in C, built specifically for high-cardinality IoT workloads — think millions of sensors each with their own time series. It bundles stream processing, built-in caching, and an AI forecasting component (TDgpt) into one system rather than requiring you to stitch together Kafka, InfluxDB, and a separate ML pipeline. The target audience is engineers running industrial or vehicle telemetry at scale who want to avoid managing five different infrastructure components.
The 'one table per device' (supertable) model is genuinely clever — it sidesteps the high-cardinality problem that kills InfluxDB at scale by partitioning storage per time series rather than across a shared tag index. Compression is serious: TSZ for floating-point columns is a purpose-built lossy/lossless algorithm that outperforms generic codecs on sensor data. The RAFT-based clustering is built-in from the start, not bolted on; sharding and replica placement are first-class concepts. The Python-based test framework in tests/ with real server integration (not mocks) is a good sign for a C database project that could easily have no tests at all.
AGPL-3.0 is a dealbreaker for any commercial product that embeds this as a library — many teams will hit that wall and only then discover TDengine Cloud exists. Windows support is described as 'limited' in their own README, which is a hard stop for any shop running Windows-based SCADA or OT systems, which is a significant chunk of the IIoT market they're targeting. The build process requires downloading external dependencies at cmake time with BUILD_CONTRIB=ON; in air-gapped industrial environments this is a real problem. The enterprise/community split is murky — the open-source tree has placeholders for S3, Azure, and COS storage that are silently disabled for community builds, so you won't know what you're missing until you hit a production limitation.