Apache Doris Boosts Real-Time Analytics
Apache Doris 4.x deepens its integration with AI-driven search and real-time query performance with a Flink Connector. This allows seamless integration of high-throughput streams (AIS, satellite, sensor) into analytical queries, enabling parallel, low-latency data retrieval and continuous synchronization for multi-modal maritime intelligence.
Apache Doris, initially known as Palo, originated within Baidu to support its ad reporting. It was open-sourced in 2017 and became an Apache Top-Level Project in June 2022. Now, its community includes over 700 contributors from various companies. The database is known for its high query speed, delivering results in sub-seconds on large datasets, and supports both high-concurrency point queries and high-throughput complex analysis. It's compatible with MySQL, enabling integration with various BI tools. Apache Doris excels in real-time data analysis, powering applications like user behavior analysis, A/B testing, log analysis, and e-commerce order analysis. Companies like TikTok, Tencent, and NetEase use it in production. JD.com processes over 10 billion rows daily with Doris, achieving 10,000+ QPS and 150ms average query latency. The Flink-Doris Connector facilitates real-time data loading from sources like Kafka or MySQL, with Flink CDC enabling full database synchronization from operational databases. Benchmarks like RTABench and JSONBench show Doris outperforming other databases in real-time analytics and JSON data handling. In RTABench, Doris was up to 6 times faster than ClickHouse and 30 times faster than PostgreSQL. In JSONBench, Doris was about 2 times faster than Elasticsearch, about 160 times faster than MongoDB, and about 1,070 times faster than PostgreSQL.
