May 2026 Summaries

One Trading, a regulated European derivatives exchange, operates under MiFID II and MiCAR, offering round-the-clock trading for crypto and equity markets with sophisticated margining solutions. They became the first to introduce 24/7 central limit order book trading for equity futures, supported by a robust matching engine capable of processing 1.8 million orders per second with sub-200 microsecond latency, thanks to collaboration with AWS on cloud-native colocation. Facing increasing trading volumes, One Trading transitioned from using Amazon DynamoDB and RDS to QuestDB, a high-performance database that allows real-time querying and efficient ingestion of over 5 million rows per second. QuestDB's infrastructure includes a distributed architecture with multi-AZ deployment and automatic failover, ensuring 99.9% uptime and seamless scalability. By self-hosting QuestDB Enterprise, One Trading achieved predictable costs and data sovereignty, enhancing their ability to scale and innovate without vendor lock-in. They acknowledged trade-offs such as incomplete AWS deployment documentation and evolving indexing, which were addressed with direct support from QuestDB's engineering team.

QuestDB 9.4.0 introduces a range of enhancements designed to optimize time-series database performance, particularly for demanding workloads such as trading floors and mission control. This version features a new, more efficient index type for SYMBOL columns, which decreases index file size and increases lookup speed. The release also includes cross-column FILL(PREV) syntax to streamline data sampling processes and an improved Web Console that facilitates the creation of materialized views and offers context-aware SQL autocompletion. Additionally, QuestDB 9.4.0 brings new window functions and text visualization capabilities, as well as various performance improvements and bug fixes across SQL planning and execution. The update aims to provide users with faster data processing, enhanced query capabilities, and improved usability, making the platform more versatile and efficient for handling extensive datasets.

QuestDB is an open-source time-series database designed for high-performance workloads, such as those found on trading floors and mission control, offering ultra-low latency and high ingestion throughput. It introduces the WINDOW JOIN syntax for efficiently aggregating data over specified time windows, significantly simplifying and speeding up complex SQL operations that involve time-based joins. This feature leverages data-level parallelism and SIMD (Single Instruction, Multiple Data) paths for fast aggregation, outperforming other databases like Timescale, DuckDB, and ClickHouse in benchmarks. QuestDB's storage layout optimizes queries by maintaining timestamp order, enabling efficient binary searches, and reducing computational overhead. The database's architecture allows it to execute queries significantly faster than its competitors, using a dedicated operator that handles data-level parallelism and vectorized processing, thus making it well-suited for temporal operations and market data analysis.

QuestDB, an open-source time-series database, is designed for high-demand environments like trading floors and mission control, offering ultra-low latency and high ingestion throughput with a multi-tier storage engine. It supports Parquet and SQL, ensuring data portability without vendor lock-in. To enhance data visualization directly within SQL result sets, QuestDB has integrated functions such as bar() and sparkline() for creating Unicode-based visualizations, allowing users to view data trends and shapes without external tools. These visualizations are tailored to the needs of traders, with functions like ohlc_bar() for candlestick charts and depth_chart() for order book profiles, making data insights more intuitive and immediate. These features keep the data exploration and debugging process efficient by reducing the need for context switching to external platforms. QuestDB is exploring additional visualization features such as heatmaps, bullet charts, and range indicators to further enhance real-time data interpretation and decision-making.