Next generation database
for capital markets
QuestDB is a high-performance, open-source time-series database for capital markets. Stream data in real-time, then offload to open formats like Parquet. Leverage extended time-series SQL for near-instant queries.
Powering critical market infrastructure at leading companies globally
Hyper-fast, reliable, open source and cost effective
Market data meets open formats
QuestDB powers the next generation of capital market infrastructure.
Live FX Orderbook Dashboard
Fast ingest and egress
Write over 8M rows/sec per server, even through dedup and out-of-order indexing.
Built for trading
Nanosecond timestamps, order book analytics, and post-trade analysis.
Open formats
No vendor lock-in. Open protocols and open source. Always.
Operational simplicity
Start in minutes with Python and SQL. No consultants needed.
Hardware efficiency
You won't need terabytes of RAM to process tick data.
Scale-ready
Scale vertically and horizontally with HA and multi-writes.
Trading technology
Low-latency exchange for Digital Assets
First MiFID II exchange in the industry achieves sub-200 microsecond round-trip latency
8 Million
Rows per second ingestion throughput
"QuestDB is an essential part of our trading platform—giving us a high-speed, scalable store for billions of trades that we can query in real time."
Steven Harper
Chief Security Officer, One Trading
Elevating trading analytics
Best Trading
Analytics Platform
QuestDB wins Best Trading Analytics Platform at the TradingTech Insight Awards Europe 2026.
Accelerate your strategies
Optimized for capital markets
Get to market, faster, with less effort.
Exchange surveillance
Capture every message flowing into exchanges with high reliability and speed.
Learn moreQuestDB Enterprise
Built for production.
Designed for scale.
99.9% uptime
High Availability, Auto failover
Security & governance
Access control (SSO/RBAC) & audit logs
Elastic scale
Multi-tier storage; storage/compute separation
Trusted partnership
SLA-backed creator support
Signals from every tick
Pre-trade analysis
Real-time indicators in SQL. SMA, EMA, VWAP, and volatility directly over tick data via window functions.
Tick-to-candle pipelines. Materialized views stream OHLCV and cascade into higher timeframes, refreshed incrementally.
Live order-book analytics. Depth, spread, and BBO lookups via
l2price()over array-stored books.Quant-stack handoff. Zero-copy exports to Pandas, Polars, DuckDB, Rust, C++, and Java for signal research.
Materialized views
Build k-line charts and cascade/chain
materialized views
Chain raw ticks to OHLCV to VWAP with incremental refresh.
No batch jobs, no stale views.
Cascade views
Chain MVs: raw ticks to 1s OHLCV to 1m OHLCV to VWAP. Each step refreshes incrementally, never full-rebuilds.
Dedup on write
Late-arriving ticks and replayed feeds stay correct without rebuilding or double-counting.
Millisecond freshness
Views refresh as new data lands. Downstream charts, alerts, and quants never wait for batch.
Downsample to Parquet
Tier view data to Parquet on disk or object storage. Queryable from QuestDB or your data lake, no duplication.
QuestDB brings a SQL-first solution that is easy to connect to our existing tools and able to ingest data at speed. Their commitment to open standards and close collaboration gives us confidence as we continue building out our data capabilities.
Denys Rtveliashvili
Head of Data, Laser Digital (Nomura Group)
9.7/10 G2 Performance RatingDeploy market leading workloads at scale.
QuestDB offers massive performance, hardware efficiency and open formats. Save costs, avoid vendor lock-in and lengthy onboarding processes, and move fast with our expert team.
HORIZON JOIN
Post-trade analysis
Measure execution quality across every fill, venue, and counterparty. Run full TCA pipelines from slippage and markout curves to implementation shortfall decomposition. Score venues and liquidity providers, benchmark against mid, BBO, and VWAP, and generate best execution reports at scale. Built for trading desks, execution algos, and compliance teams.
-- sell side markoutDECLARE @mid := (m.best_bid + m.best_ask) / 2SELECT t.symbol, h.offset / 1000000000 AS horizon_sec,count() AS n,avg((t.price - @mid) / t.price * 10000) AS a_markout_bps,sum((t.price - @mid) * t.quantity) AS total_pnlFROM fx_trades tHORIZON JOIN market_data m ON (symbol)LIST (-1s, 0, 1s, 5s, 10s) AS hWHERE t.side = 'sell' AND t.timestamp IN '$now-1h..$now'GROUP BY t.symbol, horizon_secORDER BY t.symbol, horizon_sec;
As-of Join
Match every fill to its prior book snapshot. Powers slippage vs mid and top-of-book.
Window Join
Aggregate quotes in pre/post bands around every fill. Drives implementation shortfall and pre/post markouts.
Horizon Join
Markouts at every horizon in one query, per counterparty, venue, or strategy.
High-Performance data transfer for backtesting
Backtesting made
simple
Extract historical market data through native Apache Parquet and integrate seamlessly with programming languages such as Rust, Python, C++, and Java.
Execute complex, multi-pass analyses and random data access to develop and test sophisticated strategies through popular data processing tools like Polars or Pandas.
Process every exchange message
QuestDB for
exchanges
QuestDB enables comprehensive exchange surveillance by capturing every message with high throughput, allowing for real-time monitoring and detection of suspicious activities. Its efficient data storage ensure that all transactional data is preserved in compliance with regulatory requirements, facilitating accurate reporting and audits.
Comprehensive data storage
Store every message: New orders, Acks, Fills, Matching engine outputs, Market data, Post‐trade and settlement, and network traffic data. Don't miss a single event.
Latency reporting
Build mission-critical latency reports about end-to-end travel of each message and detect anomalies in real-time. Every millisecond is crucial.
Real-time surveillance
Monitor trading activity in real-time, detect suspicious activity, and build reports for regulators. Uncompromising reliability and performance when under observation.
Many databases are easy to use when you first start them up. The difference is, QuestDB works just as well many months (and terabytes) later.
Maxwell Becker
Lead System Engineer, Anti Capital
A single time-series engine connecting agents, apps, and systems to your data.
Hyper ingestion, microsecond queries, and powerful SQL.
Lower bills through peak efficiency.