10 Best Time Series Databases of 2026

1. 1

   TimescaleDB on Tiger Data

   A managed PostgreSQL cloud built around the TimescaleDB extension for time series and real-time analytics.

   8.9overall

   A managed PostgreSQL cloud built around the TimescaleDB extension for time series and real-time analytics.

   Best forTeams that want time series tables, compression and SQL joins in the same PostgreSQL database as their application data.

   Adoption data

   2,000Tiger Data, the company formerly called Timescale, said when it renamed itself that it serves 2,000 customers and has raised $180 million.

   Tiger Data, Building Modern PostgreSQL (2025): “Since then, we have built a thriving business: 2,000 customers, mid 8-digit ARR (>100% growth year over year), $180 million raised from top investors.”

   Tiger Data, the company that was Timescale, runs a managed PostgreSQL cloud built on the TimescaleDB extension. The appeal for time series work is that hypertables, compression and more than 200 time-based SQL functions live inside ordinary Postgres, so sensor readings and metrics can be joined with customer, device and billing tables without a second database. The pricing page lists two self-serve plans: Performance from $30 per month for compute with storage at $0.177 per GB-month, and Scale from $36 per month with storage at $0.212 per GB-month, both billed hourly. Storage prices assume 5x compression, which is the company's stated customer average. Performance allows up to 8 CPU and 32 GB memory per service, while Scale raises that to 32 CPU and 128 GB and adds tiered storage and read replicas. The company reports 2,000 customers. There is no permanent free cloud tier, only a 30-day trial. We compared documented features and published prices.

   Strengths

   • +Time series and relational data share one database and one SQL dialect
   • +Published hourly compute and per-GB storage prices
   • +Runs on AWS with an Azure option

   Limitations

   • −No permanent free cloud plan, only a 30-day trial
   • −Performance plan caps services at 8 CPU and 32 GB memory
   • −Storage is billed separately from compute

   SummaryTimescaleDB on Tiger Data is the best all-round time series database for teams already comfortable with PostgreSQL. Budget for storage on top of the $30 compute price.

2. 2

   ClickHouse Cloud

   A managed service for ClickHouse, the open source columnar database for real-time analytics over very large event and time series tables.

   8.8overall

   A managed service for ClickHouse, the open source columnar database for real-time analytics over very large event and time series tables.

   Best forTeams running analytical queries over billions of time-stamped events, logs and metrics.

   Adoption data

   3,000ClickHouse said in January 2026 that more than 3,000 customers use its managed ClickHouse Cloud service.

   ClickHouse, ClickHouse raises $400M Series D, acquires Langfuse, launches Postgres (2026): “The company now serves more than 3,000 customers on its fully managed service, ClickHouse Cloud, with ARR growing more than 250 percent year over year.”

   ClickHouse is a columnar analytics database rather than a dedicated time series store, but its speed on wide, append-heavy tables makes it a common home for events, logs, metrics and market data. ClickHouse Cloud meters compute and storage separately and scales services to zero when idle. In AWS us-east-1, Basic compute costs $0.2181 per unit per hour, Scale $0.2985 and Enterprise $0.3903, and storage is $25.30 per TB per month on every tier. New accounts get $300 of credit for 30 days. Scale adds private networking, S3 role-based access and automatic vertical scaling, while Enterprise adds customer-managed encryption keys and HIPAA and PCI compliance. ClickHouse said in January 2026 that more than 3,000 customers use the managed service, and 3.6% of 2026 Stack Overflow database respondents had worked with ClickHouse. The tradeoff is cost predictability: hourly compute plus storage needs modeling before launch. We compared documented features and published prices.

   Strengths

   • +Columnar engine handles very large time-stamped tables
   • +Idle services scale to zero, which cuts cost for bursty use
   • +Published per-hour compute and per-TB storage prices

   Limitations

   • −No permanent free plan, only $300 of trial credit
   • −Usage pricing makes monthly bills harder to predict
   • −Basic tier limits storage and memory

   SummaryClickHouse Cloud is the pick when time series queries scan billions of rows and need sub-second answers. Model compute hours before you move production data.

3. 3

   InfluxDB

   A purpose-built time series database from InfluxData, offered as open source Core, self-managed Enterprise and managed cloud.

   8.8overall

   A purpose-built time series database from InfluxData, offered as open source Core, self-managed Enterprise and managed cloud.

   Best forEngineers running metrics, IoT and telemetry pipelines who want a dedicated time series engine with an open source option.

   Adoption data

   3.6%In the 2026 Stack Overflow Developer Survey, 3.6% of the 12,061 respondents to the database question had done extensive development work in InfluxDB over the past year.

   Stack Overflow, Databases, Technology data 2026, Stack Overflow Developer Survey (2026): “InfluxDB 432 3.6%”

   InfluxDB is the best known purpose-built time series engine, and InfluxData now sells it in several forms. InfluxDB 3 Core is free open source software for local development and small self-managed setups, without high availability. InfluxDB 3 Enterprise adds high availability, read replicas and SSO, priced by CPU configuration with a 30-day trial. In the cloud, Cloud Serverless bills usage at $0.0025 per MB written, $0.012 per 100 query executions, $0.002 per GB-hour of storage and $0.09 per GB out, with $250 of free credit, while Cloud Dedicated gives single-tenant clusters on custom pricing. AWS also resells it as Amazon Timestream for InfluxDB. The Telegraf collector, with more than 400 plugins, makes it easy to pull metrics from servers and devices. In the 2026 Stack Overflow survey, 3.6% of database respondents had worked with InfluxDB, the highest share of any time series specialist. The main drawback is choice overload across editions. We compared documented features and published prices.

   Strengths

   • +Free open source Core plus a 30-day Enterprise trial
   • +Usage prices for writes, queries, storage and egress are published
   • +Available managed on AWS, Azure and GCP

   Limitations

   • −Cloud Serverless has service quotas and no high availability
   • −Enterprise and Cloud Dedicated need a sales quote
   • −Several product generations to choose between

   SummaryInfluxDB is the strongest dedicated time series engine for metrics and sensor data, with a real free tier. Pick the edition carefully, because Serverless lacks high availability.

4. 4

   Grafana Cloud

   A managed observability platform whose metrics service stores Prometheus and OpenTelemetry time series.

   8.6overall

   A managed observability platform whose metrics service stores Prometheus and OpenTelemetry time series.

   Best forTeams that run Prometheus for infrastructure and application metrics and want it hosted with dashboards included.

   Adoption data

   10,000+Grafana Labs says more than 10,000 customers use its observability platform, from startups to enterprises.

   Grafana Labs, About Grafana Labs (2026): “We're trusted by over 10,000+ customers, from startups to enterprises”

   Prometheus is the default open source metrics database for cloud infrastructure, but it has no paid edition of its own, so this list scores Grafana Cloud, the managed service most teams use to host Prometheus-style metrics. Grafana Cloud Metrics accepts Prometheus and OpenTelemetry data and sits alongside logs, traces, profiles and the Grafana dashboards. The free tier is permanent, with 10,000 active series and 14 days of retention. Pro costs a $19 monthly platform fee that includes 10,000 active series and 13 months of metrics retention, then $6.50 per 1,000 active series with automatic volume discounts. Enterprise starts at a $25,000 yearly spend commit. Grafana Labs says it has more than 10,000 customers. The billing unit is the catch: high-cardinality labels multiply active series and therefore cost. It also is not a general SQL store for sensor or trade analytics. We compared documented features and published prices.

   Strengths

   • +Cheapest fixed paid entry in this list at $19 per month
   • +Permanent free tier that needs no credit card
   • +Dashboards and alerting come with the metrics store

   Limitations

   • −Billed per active series, which grows with label cardinality
   • −Built for observability metrics, not general SQL analytics
   • −Enterprise starts at a $25,000 yearly spend commit

   SummaryGrafana Cloud is the simplest managed home for Prometheus metrics, with a real free tier and a $19 paid entry. Watch label cardinality, because active series drive the bill.

5. 5

   QuestDB

   An open source, SQL time series database built for low-latency market data, with a commercial Enterprise edition.

   8.3overall

   An open source, SQL time series database built for low-latency market data, with a commercial Enterprise edition.

   Best forTrading, market data and telemetry teams that need very high ingest with standard SQL.

   Adoption data

   17.5kQuestDB's open source repository on GitHub has 17.5k stars.

   GitHub, questdb/questdb (2026): “Stars 17.5k stars”

   QuestDB was started by engineers from electronic FX trading and is aimed at workloads where latency matters, such as market data, order books, risk and surveillance, with growing use in energy, aerospace and robotics. Its open source core is licensed under Apache 2.0, uses standard SQL and builds in durability, deduplication and out-of-order correction for streams that arrive late. Data can be read through Apache Parquet, Iceberg and Arrow, which keeps history usable by other tools. QuestDB Enterprise adds tiered storage, high availability with automatic failover, a 99.9% uptime target, SSO, RBAC and audit logs, and it is sold through sales with a free trial rather than a public price list, so the USD check is pending. The company says it runs at tier-1 banks, exchanges, market makers and hedge funds, and its GitHub repository has 17.5k stars. The ecosystem is smaller than PostgreSQL's. We compared documented features and published prices.

   Strengths

   • +Free open source core for self-managed use
   • +Designed for tick and order book data from capital markets
   • +Open formats keep data readable outside QuestDB

   Limitations

   • −No published USD price for Enterprise
   • −Smaller ecosystem than PostgreSQL or InfluxDB
   • −High availability and access control require Enterprise

   SummaryQuestDB is the strongest open source option for market data and other high-ingest streams that need SQL. Expect a sales conversation for high availability.

6. 6

   TDengine

   An open source time series database and industrial data platform built as a modern alternative to data historians.

   8.1overall

   An open source time series database and industrial data platform built as a modern alternative to data historians.

   Best forManufacturing, energy and industrial IoT teams that track thousands of sensor tags.

   Adoption data

   1,515.4KTDengine reports about 1.5 million instances of its open source time series database running worldwide.

   TDengine, TDengine homepage (2026): “1,515.4K Instances worldwide”

   TDengine targets industrial data: sensor tags from factories, energy sites and grids that older data historians have stored for decades. Its All-in-One product bundles the TDengine TSDB with ingestion, analytics, modeling, visualization and AI features, and it can run at the edge, in containers, on servers or in public, private or hybrid clouds. Pricing is unusually transparent. Deployments up to 5,000 tags are free, including small production use. Above that, an annual license for 7,500 tags costs $12,500 a year, or $31,250 as a perpetual license with optional annual support of $4,688. Official support for free deployments costs $3,375 a year. The company reports about 1.5 million open source instances worldwide and 24,000 GitHub stars. A recent customer story describes 35 million rows of energy data a day. For web metrics or SQL analytics outside the plant, general tools on this list fit better. We compared documented features and published prices.

   Strengths

   • +Transparent license prices published by tag count
   • +Free tier allowed for small production deployments
   • +Built for plant and grid sensor data

   Limitations

   • −Paid licenses start at $12,500 a year
   • −Official support is a separate annual fee
   • −Narrower fit outside industrial and IoT data

   SummaryTDengine is the clearest pick for industrial sensor data, with a free 5,000-tag tier and published license prices. Teams outside manufacturing and energy have better general options.

7. 7

   VictoriaMetrics Cloud

   A managed service for VictoriaMetrics, the open source Prometheus-compatible time series database for metrics.

   8.1overall

   A managed service for VictoriaMetrics, the open source Prometheus-compatible time series database for metrics.

   Best forTeams that outgrow a single Prometheus server and want a cheaper, PromQL-compatible metrics store.

   Adoption data

   1 billionVictoriaMetrics says its open source projects have reached 1 billion Docker pulls and 19 million GitHub downloads.

   VictoriaMetrics, About VictoriaMetrics (2026): “1B Docker Pulls 19M GitHub Downloads”

   VictoriaMetrics is an open source time series database that accepts Prometheus data and queries, and the company says it uses up to 80% less memory and up to 2.5x less storage than alternatives. VictoriaMetrics Cloud runs it as a managed service with fixed tiers rather than per-series billing. The Initial tier costs from $200 per month and covers 500k active time series, 16.6k datapoints per second and 170 GB of daily reads, with business-hours support. New accounts get $200 of credit for 30 days, and billing can run through AWS Marketplace. External traffic costs $0.09 per GB. The same cloud hosts VictoriaLogs and VictoriaTraces, so logs and traces can sit beside metrics. The company reports 1 billion Docker pulls and 19 million GitHub downloads across its projects. The entry price is high for a handful of servers, where the free self-hosted edition or Grafana Cloud's free tier fits better. We compared documented features and published prices.

   Strengths

   • +Tier limits make monthly cost easy to forecast
   • +Drop-in PromQL compatibility for Prometheus users
   • +Free open source edition for self-hosting

   Limitations

   • −Managed entry price of $200 per month is high for small setups
   • −External traffic billed at $0.09 per GB
   • −Focused on observability metrics, not general analytics

   SummaryVictoriaMetrics Cloud suits teams with hundreds of thousands of Prometheus series who want flat, predictable pricing. Small setups should start with the free open source edition.

8. 8

   Amazon Timestream for InfluxDB

   AWS's managed time series service that runs InfluxDB instances inside your AWS account.

   7.9overall

   AWS's managed time series service that runs InfluxDB instances inside your AWS account.

   Best forAWS teams that want InfluxDB billed to their AWS account, with IAM and VPC integration.

   Adoption data

   $46.0 billionAmazon reported AWS segment sales of $46.0 billion in the third quarter of 2026, up 39%; Amazon does not report Timestream usage separately.

   Amazon, Amazon.com Announces Third Quarter Results (2026): “AWS segment sales increased 39% year-over-year to $46.0 billion.”

   Amazon Timestream now has two flavors, and AWS steers new customers to Timestream for InfluxDB: its documentation says it recommends that new customers evaluate the InfluxDB version instead of Timestream for LiveAnalytics. Timestream for InfluxDB runs managed, highly available InfluxDB instances in Single-AZ or Multi-AZ deployments. On-demand instances are billed per instance-hour in 1-second increments with a 10-minute minimum after a status change, plus storage, and AWS credits and enterprise discount agreements apply. InfluxData lists it as its AWS-native option with IAM and VPC integration and AWS support plans. AWS's own worked example prices a Single-AZ db.influx.2xlarge with 400 GiB of storage at $737.88 a month, which shows that always-on instances are not cheap. Because the pricing tables load dynamically, our USD check of the entry price is pending. Amazon does not report Timestream usage, so adoption reflects AWS as a whole. We compared documented features and published prices.

   Strengths

   • +Billing and security stay inside the AWS account
   • +No long-term commitment for on-demand instances
   • +Uses the open InfluxDB engine and APIs

   Limitations

   • −Instance pricing adds up quickly for always-on workloads
   • −Timestream for LiveAnalytics is no longer the recommended path for new customers
   • −AWS only, with no multi-cloud option

   SummaryTimestream for InfluxDB is the easy choice for AWS-only teams that want InfluxDB without running servers. Compare its instance-hour cost with InfluxData's own serverless cloud first.

9. 9

   kdb+ by KX

   A columnar, in-memory time series database and analytics engine programmed in the q language, now extended as KDB-X.

   7.7overall

   A columnar, in-memory time series database and analytics engine programmed in the q language, now extended as KDB-X.

   Best forBanks, hedge funds and exchanges that need the lowest latency on tick data and can invest in q skills.

   Adoption data

   60%KX says financial institutions representing more than 60% of global GDP use its software.

   KX, About KX (2026): “Financial Institutions representing more than 60% of global GDP use KX”

   kdb+ has been the reference time series engine on trading floors for more than 30 years. KX describes a columnar, in-memory system smaller than 1 MB that fits in CPU cache, paired with the array-based q language, so queries scan billions of records as vectors rather than rows. Real-time and historical data live in one environment, which is why banks use it for tick capture, execution analytics and surveillance. KDB-X is the next generation, adding vector search, AI workloads and optional NVIDIA GPU acceleration, and KX publishes benchmark claims such as 34x faster average query speed against competitors. Pricing for KDB-X is set with a slider on the KX site and market data or surveillance solutions are quoted by sales, so our USD check is pending. KX says financial institutions representing more than 60% of global GDP use its software. The price of that speed is q: hiring and training take real effort. We compared documented features and published prices.

   Strengths

   • +Very low latency on streaming and historical tick data
   • +Deep track record in capital markets
   • +One engine for real-time and historical analysis

   Limitations

   • −The q language has a steep learning curve
   • −No published USD price we could verify
   • −Ecosystem is narrow outside finance

   Summarykdb+ remains the specialist choice for latency-critical tick data in finance. Most other teams will get further, faster, with a SQL time series database.

10. 10

    CrateDB Cloud

    A distributed SQL database for real-time analytics on time series, JSON, full-text, vector and geospatial data, offered as a managed cloud.

    7.6overall

    A distributed SQL database for real-time analytics on time series, JSON, full-text, vector and geospatial data, offered as a managed cloud.

    Best forIndustrial and IoT teams that want time series, search and geospatial queries in one SQL engine.

    CrateDB is a distributed SQL database from an Austrian company, aimed at live operational data from sensors, telemetry streams and applications. Its pitch is that one SQL query can mix time series aggregations, JSON, full-text search, vector similarity and geospatial joins, without pre-aggregating data or flattening schemas. It speaks the PostgreSQL wire protocol, offers a REST endpoint and has connectors for Kafka, Spark and dbt, and the core is Apache 2.0 licensed. CrateDB Cloud has a permanent free plan with 2 vCPUs, 2 GB RAM and 8 GB of storage. Shared clusters for non-critical work start at $0.073 per hour, Dedicated clusters at $0.246 per hour including basic support, and larger custom setups are quoted. A STACKIT option keeps data in Europe. We found no citable customer or usage count, so adoption is scored conservatively. We compared documented features and published prices.

    Strengths

    • +Permanent free plan with no credit card required
    • +Joins device metadata at query time without pre-aggregation
    • +European ownership and hosting options for data sovereignty

    Limitations

    • −No citable customer or usage figures, so adoption is scored conservatively
    • −Shared plan is for non-critical workloads only
    • −Smaller community than PostgreSQL or ClickHouse

    SummaryCrateDB Cloud fits teams whose time series queries also need search, JSON and location data in one SQL engine. Its smaller user base is the main risk.

Questions or corrections: editorial@statlyst.com · Disclosure