New TDengine Benchmark Results Show Up to 37.0x Higher Query Performance Than InfluxDB and TimescaleDB

TDengine Team
Share on LinkedIn

TDengine Outperforms Other Time-Series Databases Across the Board in Industry-Standard TSBS Evaluation With Up to 26.9x Less Storage and 10.6x Faster Ingestion

LOS GATOS, Calif., Feb. 28, 2023 — TDengine, the popular open-source time-series data platform, today revealed new Time Series Benchmark Suite (TSBS) performance test results for TDengine 3.0.

TSBS is an open-source, independent benchmarking framework for time-series databases that tests data ingestion and query performance based on sample datasets. On identical AWS environments, the tests revealed the significant performance advantages of the TDengine time-series data platform over TimescaleDB and InfluxDB.

  • Ingestion: TDengine wrote the TSBS datasets 1.5 to 6.7 times faster than TimescaleDB and 3.0x to 10.6x faster than InfluxDB. In all scenarios, TDengine write performance was greater.
  • Queries: TDengine outperformed TimescaleDB and InfluxDB in all queries, and the gap only widened as the complexity of queries increased.
  • In the 100-device test scenario, TDengine responded between 1.1x and 28.6x faster than TimescaleDB and between 1.9x and 37.0x faster than InfluxDB.
  • In the 4,000-device test scenario, TDengine responded between 1.2x and 24.6x faster than TimescaleDB and between 1.8x and 34.2x faster than InfluxDB.
  • Storage: TDengine’s efficient data storage and compression mechanisms enabled the benchmark dataset to occupy far less disk space than InfluxDB or TimescaleDB, especially at scale. With the largest data sets of 10 million devices, TDengine used 4.5x less storage than InfluxDB and 26.9x less than TimescaleDB.

“As time-series datasets continue their enormous growth, a high-performing system is essential for keeping response times short and operational costs low,” said Jeff Tao, founder and CEO of TDengine. “With the release of our TSBS evaluation results, enterprises, manufacturers, utilities, and other industries embracing time-series data solutions can have full confidence in their decision to deploy TDengine—regardless of the size or scope of their data needs.”

A Modern Open-Source Time-Series Database

Released as open-source software in 2019, TDengine has over 20,000 stars on GitHub and over 190,000 installations across 50 countries worldwide. TDengine combines a database with caching, stream processing, and data subscription as a complete, purpose-built solution for time-series data.

In 2022, the company unveiled TDengine 3.0, delivering a fully distributed architecture that decouples compute and storage resources for dynamic scaling and deployment across public, private, and hybrid clouds. Later that year, it also released TDengine Cloud, a fully managed time-series database solution that delivers the industry-leading performance of TDengine 3.0 as a cloud service. Recently, it revealed the TDengine PI Connector, further expanding the TDengine data platform to help industrial users process time-series data efficiently.

Methodology and Benchmark Details

TSBS is an open-source, independent benchmarking framework for time-series databases that tests data ingestion and query performance based on sample datasets. The framework has been used by leading time-series DBMS vendors since 2018 and continues to be the most widely supported platform for testing time-series database performance.

TDengine’s evaluation compares TDengine 3.0, TimescaleDB 2.6, and InfluxDB 1.8 running on identical AWS environments with the TSBS DevOps (CPU-only) dataset. For detailed information about all tests performed and steps to reproduce, see the full report.

About TDengine

TDengine is a popular open-source data platform purpose-built for time-series data. With over 20,000 stars on GitHub and hundreds of new installations daily, TDengine is used in more than 50 countries worldwide. The company is headquartered in Los Gatos, CA and has raised $70M in venture capital.

Press Contact

For press or business inquiries, contact us.