QuestDB — High-Performance Time-Series Database with SQL

What it is

QuestDB is a fast open-source time-series database built in Java and C++. It ingests millions of rows per second via the InfluxDB line protocol and lets you query with standard SQL extended with time-series functions like SAMPLE BY, LATEST ON, and ASOF JOIN. QuestDB is column-oriented and optimized for append-heavy workloads typical of IoT, financial, and observability data.

QuestDB is designed for engineers building applications that process high-volume time-stamped data: metrics, sensor readings, financial ticks, or application logs.

How it saves time or tokens

Traditional databases struggle with high-frequency inserts and time-range queries over large datasets. QuestDB's column-oriented storage and memory-mapped files provide consistent query performance regardless of data volume. The SQL interface means no new query language to learn. InfluxDB line protocol compatibility lets you point existing Telegraf or vector.dev pipelines at QuestDB without changing your collection layer.

How to use

1. Start QuestDB with Docker:

docker run -d --name questdb \
  -p 9000:9000 -p 9009:9009 -p 8812:8812 \
  questdb/questdb

1. Access the web console at http://localhost:9000 and run SQL queries.

1. Ingest data via InfluxDB line protocol (port 9009):

echo 'sensors,location=office temperature=23.5,humidity=45 1681000000000000000' | \
  nc localhost 9009

Or use the PostgreSQL wire protocol (port 8812) with any Postgres client.

Example

Time-series analysis with QuestDB SQL extensions:

-- Average temperature per hour
SELECT timestamp, avg(temperature) AS avg_temp
FROM sensors
WHERE timestamp > dateadd('d', -7, now())
SAMPLE BY 1h
ALIGN TO CALENDAR;

-- Latest reading per sensor
SELECT * FROM sensors
LATEST ON timestamp PARTITION BY location;

-- Join two time-series by closest timestamp
SELECT s.timestamp, s.temperature, w.wind_speed
FROM sensors s
ASOF JOIN weather w ON (s.timestamp = w.timestamp);

Related on TokRepo

• Database tools — Browse database solutions
• Monitoring tools — Explore observability and monitoring tools

Common pitfalls

• Using QuestDB for general-purpose OLTP workloads. QuestDB is optimized for append-only time-series data. It does not support UPDATE or DELETE on individual rows in the traditional sense.
• Not using SAMPLE BY for time-series aggregation. Writing manual GROUP BY with date functions is slower than QuestDB's native SAMPLE BY, which is optimized for time-bucketed queries.
• Ignoring the designated timestamp column. QuestDB requires a designated timestamp column for time-series features. Without it, SAMPLE BY, LATEST ON, and ASOF JOIN will not work.
• Starting with an overly complex configuration instead of defaults. Begin with the minimal setup, verify it works, then customize incrementally. This approach catches configuration errors early and keeps troubleshooting straightforward.