Why QuestDB
AI is rewriting data infrastructure.
We’re at its core.
Legacy data infrastructure is being rewritten because of AI, cost pressures,
and vendor lock-in. QuestDB is built for it.
Ingest & egress performance on
QuestDB v10.0
QuestDB wins Best Trading Analytics Platform at the TradingTech Insight Awards Europe 2026.
Read more
Standard SQL. no lock in.
Easy to write SQL for developers and agents
Special SQL primitives for time-series, built for the most demanding workloads in capital markets, aerospace, and energy. Easy for engineers to write, native for AI agents.
Experience QuestDB now
Explore our live demo
Our live demo instance is the quickest way to get a feel for QuestDB. Scan more than 2 billion rows in milliseconds. Example queries can get you deep in just a few clicks.
/* This query takes the best bid price and multiplies it by 1.01. Then it finds the level at which that price would be met inside the prices array. We slide the volume array from the first level until the level we found, and sum all the volumes. Note the use of DECLARE to make the query more readable. */
LLMs speak SQL.
Your database should too.
Models speak SQL and read open formats like Parquet and Iceberg. QuestDB supports both, exposed via a REST API,
so one prompt moves an agent from intent to executed query, with no proprietary client in between.
From real-time ingest to open data lake
Delete proprietary, commit open
Ingested data is offloaded into an open data lake in Parquet format, with automatic tiering. One SQL surface queries real-time and historical data across all tiers. If you're already using Python, point your dataframe libraries and AI frameworks at the Parquet files directly, no export needed.
2025-01-01T00:00:00.000
EURUSD
1.0847
2500000
2025-01-01T00:00:01.000
GBPUSD
1.2734
1800000
2025-01-01T00:00:02.000
USDJPY
149.82
3200000
2025-01-01T00:00:03.000
AUDUSD
0.6452
1200000
2025-01-01T00:00:04.000
USDCAD
1.3765
1500000
2025-01-01T00:00:05.000
USDCHF
0.8834
900000
Azure Blob
Amazon S3
NFS
One SQL engine, multiple storage tiers
Real-time and historical data, in one engine.
WHERE symbol in ('AAPL', 'NVDA')
LATEST ON timestamp PARTITION BY symbol
CREATE MATERIALIZED VIEW 'trades_OHLC'
min(price) AS low
timestamp IN today()
SELECT spread_bps(bids[1][1], asks[1][1])
FROM read_parquet('trades.parquet')
SAMPLE BY 15m
Open by Design
Open source. Open standards.
Peak performance time-series
Time-series
ingestion at scale
Write-ahead logging (WAL) for instant durability; time-partitioned columnar engine for speed. Ingest millions of rows/sec and query with SIMD-optimized SQL.
“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.”
Faster ingestion than leading competitors in open source benchmarks (TSBS)
Plug QuestDB into the tools
your team already uses.
Native integrations across the modern open data stack.
Real-time analytics for high performance workloads
Built for the most demanding
time-series use cases
Need a hand?
Join our developer community
Learn how other teams get the most out of QuestDB and participate in our development.
QuestDB community forum