QuestDB

What is QuestDB?

QuestDB is an open source time-series database built for high-throughput ingestion and fast SQL queries over huge volumes of timestamped data. It ingests millions of rows per second, stores them in memory-mapped columnar partitions, and lets you query with standard SQL plus time-series extensions like SAMPLE BY and ASOF JOIN.

What is QuestDB best for?

QuestDB is best for workloads where data arrives fast and queries need to be near-instant: capital markets and crypto trading, IoT and sensor telemetry, energy grid monitoring, application and infrastructure metrics, and real-time dashboards. It suits teams who already know SQL and want a single, fast engine instead of stitching together a separate ingestion pipeline and analytics store.

What can QuestDB do?

• Ingest data over the InfluxDB Line Protocol, a REST HTTP API, or PostgreSQL wire protocol — benchmarked above 19M rows/second per server.
• Query with standard SQL extended for time series: SAMPLE BY time-bucketing, LATEST ON, ASOF JOIN and WINDOW JOIN for point-in-time matching, plus materialized views.
• Connect existing tools through PostgreSQL wire compatibility (Grafana, Superset, psql) and Python clients (Pandas, Polars).
• Stream data in from Kafka, Redpanda, Telegraf, and Apache Flink.
• Store partitions in native columnar format or convert to Apache Parquet, with an interactive web console for writing queries.
• Run on Docker, Kubernetes, Linux, macOS, or Windows.

Where does QuestDB fall short?

• It’s a specialized append-oriented time-series store, not a general-purpose relational database. There’s no rich transactional OLTP support, and update/delete semantics are limited — if you need a normal application database with joins across many mutable tables, PostgreSQL or ClickHouse may fit better.
• Several production-grade features are Enterprise-only: high-availability replication, role-based access control, TLS/SSO, and automatic tiering to object storage. The Apache-2.0 open source build runs as a single node without these.
• The engine runs on the JVM (with C++ and Rust hot paths), so memory-mapped columns and large working sets can want substantial RAM for big datasets compared with a lighter embedded store.

Is QuestDB free?

Yes — the core database is free and open source under the Apache-2.0 license, and you can self-host it at no cost with no row, retention, or connection caps. QuestDB Enterprise (self-managed) and QuestDB Cloud (fully managed on AWS, Azure, or GCP) are paid tiers that add replication, security, object-storage tiering, and support.

What does QuestDB replace?

QuestDB is an open source alternative to managed and proprietary time-series and observability products. Teams use it in place of Amazon Timestream for serverless time-series storage, and for metrics and telemetry workloads it can stand in for parts of Datadog and Splunk — keeping high-cardinality data on infrastructure you control rather than paying per-host or per-GB ingestion fees.

FAQ

Is QuestDB open source? Yes. The core database is licensed under Apache-2.0, with 170+ contributors and the full source on GitHub. Enterprise and Cloud are commercial editions built on top of it.

Can I self-host QuestDB for free? Yes. The open source build runs free on Docker, Kubernetes, or a plain VM with no usage limits. You only pay if you choose Enterprise or the managed Cloud service.

Is QuestDB a good alternative to Amazon Timestream? For many time-series workloads, yes — QuestDB offers standard SQL, much higher benchmarked ingestion throughput, and no per-write cloud pricing. The tradeoff is that you run and scale it yourself unless you use QuestDB Cloud.

What do I need to run QuestDB? A machine with a Java runtime (bundled in the official Docker image and binaries) and enough RAM for your working set. It runs on Linux, macOS, and Windows, and ships Docker images and Kubernetes Helm charts.