The Top Time-Series Databases
This is a list of the top commercial, financial and open source time-series databases available as of January 2023.
What is a time-series Database?
A time-series database is specialized to quickly and efficiently answer queries involving:
- Time-Joins - e.g. Event X occurred at 9am, when was the closest Y event to that.
- Time-Aggregations - specialized functions to allow handling date/time types well.
- Compression - As the data is large and often repeating
- Support Nanoseconds - As for some industries e.g. trading, the exact timing of events matter.
The table below shows the support each database has in this area.
Time-series Databases
Product | Score | Released | Speed | SQL | Compression | Time-Joins | Time-Aggregations | Nanoseconds | Popularity | Description | License |
---|---|---|---|---|---|---|---|---|---|---|---|
Timescale (wp) | 6 | 2018 | Medium+ | Full + Extensions | Basic | No> | Micros | Unknown | Postgres for time-series | Apache License 2.0 | |
Clickhouse (wp) | 8 | 2016 | Fast | Some + Custom | Many | asof | Partial | Popular | Very fast OLAP database with cloud version available. Started 10 years ago at Yandex to store the russian equivalent of google analytics. | Apache License 2.0 | |
QuestDB | 7 | 2014 | Fast | High + Extensions | Many | asof+ | Good | Partial | New | Fast database with strong focus on time-series. Very similar ideas to kdb+ but open source. | Apache License 2.0 |
InfluxDB (wp) | 6 | 2013 | Medium | Some + Custom | Some | No | Yes | IoT/monitoring | Originally built for monitoring and alerting. Now specializing in time-series analysis and IoT. Uses an SQL-like language. | MIT License | |
Druid (wp) | 6 | 2011 | Medium | Some + custom | No | Milliseconds | Click analytics | A distributed data store written in Java. Druid is designed to quickly ingest massive quantities of event data, and provide aggregated queries ontop. | Apache License 2.0 | ||
kdb+ (wp) | 8 | 2003 | Fast++ | Some + qSQL | Many | Yes. AJ/WJ | Finance | Very fast column-oriented database with custom language q and custom time-series joins.
Steep learning curve and difficult to find experts. |
Commercial |
Time-Series Benchmarks
For more information see our Time-Series benchmarks article
Clickbench results:
System & Machine | Relative time (lower is better) | Note |
---|---|---|
ClickHouse (c6a.metal, 500gb gp2): | ×1.59 | |
SelectDB (c6a.metal, 500gb gp2): | ×1.88 | |
ClickHouse (m5d.24xlarge): | ×2.15 | |
StarRocks (c6a.metal, 500gb gp2): | ×2.16 | |
Redshift (4×ra3.16xlarge): | ×2.20 | |
DuckDB (c6a.metal, 500gb gp2): | ×2.74 | |
QuestDB (partitioned) (c6a.metal, 500gb gp2)†: | ×24.24 | |
MariaDB ColumnStore (c6a.4xlarge, 500gb gp2)†: | ×59.27 | |
TimescaleDB (compression) (c6a.4xlarge, 500gb gp2): | ×86.91 | |
Druid (c6a.4xlarge, 500gb gp2)†: | ×150.50 | |
PostgreSQL (c6a.4xlarge, 500gb gp2): | ×883.89 | NOT column oriented |
Results reproduced from Mark Litwintschik's excellent article.
Setup | Total Query Time (lower = better) | Note |
---|---|---|
kdb+/q & 4 Intel Xeon Phi 7210 CPUs | 1.04 | |
ClickHouse, 3 x c5d.9xlarge cluster | 4.06 | |
Clickhouse on DoubleCloud, s1-c32-m128 | 5.77 |
Financial Tick Databases
Product | Vendor (release year) | Description |
---|---|---|
One Tick Database | Onetick
2005 |
Column/Row oriented database targeted at the financial sector and specialised for tick data, created by Leonid Frants that had built a tick solution while at Goldman Sachs. |
eXtremeDB | McObject
2001 |
A fast embedded, mostly in-memory database targeted for financial firms and time series data. It's raw API and ability to be embedded within a process makes it fast, however this means a higher configuration cost and learning curve to get started. |