A New Data Model for Time Series Data Processing

Explore a groundbreaking data model for time series data processing in this 39-minute conference talk by Jeff Tao from TDengine. Delve into the innovative "one table for one data collection point" approach, which builds upon the relational data model while addressing its limitations. Learn how this new model significantly enhances data ingestion rates, query performance, and storage efficiency compared to traditional tag set and relational models. Discover the mechanics behind its superior data compression ratio, faster ingestion rates, and reduced query latency. Examine the challenges associated with managing large-scale deployments involving millions or even billions of tables, and understand how the concept of "supertables" offers a solution. Gain insights from benchmark results based on the TSBS test suite, demonstrating the effectiveness of this new data model in real-world scenarios.