InfluxDB alternatives: 11 time series databases tools compared
Teams look for an InfluxDB alternative when the Serverless meter becomes hard to forecast, when the free tier's 30-day retention runs out, or when production features turn out to sit in tiers priced only by contact. Others have standardised on Prometheus, SQL or Postgres and want one query language across tools. Open source engines, hourly cloud services and metrics stores each answer a different part of that, and we list eleven.
Short answer: Looking for a replacement for InfluxDB? QuestDB, GreptimeDB and VictoriaMetrics come first in our time series databases ranking. We open the ranking with smaller vendors and put the established brands from fourth place onward. Start with QuestDB if you want plain SQL with time-series extensions and a free engine to start on. All 11 alternatives below are ranked products from Best Time Series Databases in 2026, and each says where it differs from InfluxDB.
Vendors can pay for more visibility on this website, for example a higher place in a ranking or more room on a page. Commercial relationships can also affect which products we feature and how prominently. Every vendor link is marked the same, whatever the arrangement behind it. How that works.
Pricing: Usage-based cloud with a free tier; Dedicated and Enterprise custom-priced
Established in: United States
Built for: Small teams wanting a cloud database with a visible, itemised price list
Not for: Buyers who need high availability without speaking to sales
01
Why buyers look past InfluxDB
01
Pricing is a set of four meters
On Cloud Serverless, InfluxData prints $0.0025 per MB written, $0.012 per 100 query executions, $0.002 per GB-hour stored and $0.09 per GB sent out. Each bill is a sum of four moving quantities, so a dashboard refreshed every few seconds or a change in retention shifts the total without any change in the data you collect. Forecasting takes a model, not a plan price.
02
Higher tiers are priced by contact
Core is a single-node engine with community support and no high availability. Enterprise, sold per CPU configuration, adds clustering, read replicas, SAML and a support SLA, and Cloud Dedicated is priced on CPU, RAM and storage. None of these carries a figure on the page we read, so a production budget starts with a sales conversation, not a calculator.
03
The free tier is thin by design
The free Serverless tier keeps data for 30 days and limits ingestion rates, with quotas applying and no high availability. Paid plans unlock unlimited retention and higher rate limits. The pricing page gives no numeric ingestion cap, so a team can only find the ceiling by hitting it. Anyone who needs history longer than a month has to move to a paid meter.
dashboards must stay fast over a heavy stream and you will run a cluster yourselves
#12
03
What each one does differently
1
QuestDB
Not stated by the vendor · On request
QuestDB ships an Apache 2.0 edition to download and sells an Enterprise edition through its sales team. Its SQL adds SAMPLE BY for bucketing and ASOF JOIN for matching events by timestamp, and older data can be moved out to Parquet and Iceberg files. High availability, replication, role-based access and single sign-on sit in Enterprise, and the site shows no price.
Switch from InfluxDB if you want plain SQL with time-series extensions and a free engine to start on.
GreptimeDB is developed by GREPTIME INC, which says it began in 2022. The core runs in your own environment under Apache-2.0, while the managed Enterprise plan is billed monthly on what you store, with volume discounts. A bring-your-own-cloud arrangement is priced after a proof of concept, and the pricing page points to a live demo rather than a trial.
Switch from InfluxDB if a monthly storage-based bill and a self-run Apache-2.0 core both appeal.
VictoriaMetrics divides its range into an open source set covering metrics, logs and traces, plus paid Enterprise, managed Cloud and Anomaly Detection products. A customer quote on its home page describes a test system running within twenty minutes and calls it a drop-in Prometheus replacement. The pages we read give no prices and no licence name.
Switch from InfluxDB if your data is infrastructure metrics and you already speak the Prometheus protocol.
CrateDB comes from Austria and offers a shared entry plan, an hourly Dedicated plan with service levels and a Custom plan on contact. The core is Apache 2.0, and it can run on-premises, at the edge or as a managed cluster on STACKIT. The entry plan's limits are stated inconsistently on the pricing page, so confirm them before relying on either figure.
Switch from InfluxDB if a European vendor and an on-premises or STACKIT deployment option matter.
Tiger Data is the trading name of Timescale, Inc., creators of TimescaleDB. Performance and Scale plans are billed hourly for configured compute and storage, Scale raises the ceiling from 8 to 32 CPU per service and recovery stretches from 3 to 14 days. SAML sign-on and HIPAA wait for Enterprise. Entry is a 30-day trial.
Switch from InfluxDB if your team prefers to keep working in Postgres-style SQL on a service billed by the hour.
TDengine prices by tag count: self-hosted use is free up to 5,000 tags, after which annual or perpetual licences apply, and high availability adds thirty per cent. Managed cloud packages run on AWS, Azure or GCP and bill monthly. The site shows connectors for OPC, MQTT and Kafka but names no legal entity or country.
Switch from InfluxDB if industrial sensor feeds over OPC or MQTT make a per-tag price easy to predict.
Prometheus is a community project that keeps series locally on each server, identifies them by metric name and key-value pairs, and queries them in PromQL. It runs free under the Apache 2 licence, alerts need the separate Alertmanager, and nobody sells support for it. It collects monitoring data rather than serving as a general-purpose store for application records.
Switch from InfluxDB if the workload is monitoring and alerting, and you can run the servers yourselves.
Mimir holds Prometheus metrics for the long term, replicated across object storage such as S3 or Azure Blob. The licence is AGPLv3, which matters to anyone embedding it in a product. Grafana Labs offers it self-run, inside Grafana Cloud or as supported Enterprise Metrics, and the cloud pricing page lists metrics plans without naming Mimir itself.
Switch from InfluxDB if you need years of Prometheus history held on cheap object storage.
ClickHouse, Inc. is a Delaware corporation selling cloud plans on AWS, GCP and Azure, alongside an Apache 2.0 server you host. It is a general columnar analytics database, so you design the tables and sort keys yourself. The Basic tier caps storage at 1 TB in one availability zone, and bring-your-own-cloud is sold by contact only.
Switch from InfluxDB if time-stamped events are one workload among many analytical queries you run.
Apache IoTDB is a project of the Apache Software Foundation, presented as an industrial IoT time series database with an edge-cloud design. It claims high-speed writes from millions of low-power devices and a high disk compression ratio, and links to Hadoop, Spark, Flink and Grafana. The pages we read describe no plan, price or hosted service.
Switch from InfluxDB if fleets of connected devices write at the edge and you want a foundation-governed codebase.
Apache Druid describes itself as a real-time analytics database, not a time series database, returning sub-second queries over streaming and batch data. Ingestion attaches to streaming platforms without a separate connector, and configurable tiering trades price against speed. The site lists no hosting and no paid plan, so operation falls to your own staff.
Switch from InfluxDB if dashboards must stay fast over a heavy stream and you will run a cluster yourselves.
You mostly need Prometheus-compatible monitoring storage
VictoriaMetrics: The vendor calls it a drop-in Prometheus replacement, with an open source set and paid Enterprise and Cloud products on top.
02
You want an Apache 2.0 engine with SQL for time series
QuestDB: SAMPLE BY and ASOF JOIN are built into its SQL, and older data can move to Parquet and Iceberg.
03
You prefer a European vendor you can also run on-premises
CrateDB: It was founded in Austria, has an Apache 2.0 core and can run on a managed cluster on STACKIT.
04
You want an hourly bill and Postgres-style SQL
Tiger Data: Performance and Scale plans are billed hourly, and entry is a 30-day trial.
05
You collect from industrial equipment and want a per-tag price
TDengine: Self-hosted use is free up to 5,000 tags, with connectors for OPC, MQTT and Kafka listed on the site.
04
When InfluxDB is still the right choice
InfluxDB stays the sensible pick when you want a vendor that publishes its cloud rates and offers several ways to run the same engine: free Serverless, open source Core on Linux, Mac or Docker, Dedicated single-tenant, Enterprise on your own infrastructure, or Amazon Timestream for InfluxDB billed through AWS. Teams with Telegraf already collecting their data keep a familiar pipeline. Staying makes most sense when the metered bill is predictable at your volume.
05
What moving off InfluxDB involves
01
Queries are rewritten, not copied
The InfluxDB 3 documentation lists SQL and InfluxQL, and Flux appears only under the InfluxDB 2 products. Any Flux tasks and dashboards written against version 2 need rewriting for a SQL-based or PromQL-based target. Export a sample of your real queries first and time them on the new engine.
02
Collection and retention settings need a plan
Telegraf agents can usually keep running with a different output plugin, but retention, downsampling and alert rules live in the database and must be recreated. Run both stores side by side for a retention period so that dashboards can be compared before the old one is switched off.
06
Questions about replacing InfluxDB
6 answers
What does InfluxDB cloud cost per month?
The Serverless usage plan has no minimum commitment and charges $0.0025 per MB written, $0.012 per 100 query executions, $0.002 per GB-hour stored and $0.09 per GB out. Dedicated and Enterprise are priced by contact, so no monthly figure exists for them on the page.
Is there a free InfluxDB option that keeps data longer than 30 days?
Core is free with no licence required, runs on a single node and has community support only. The free Serverless tier keeps 30 days.
Alternatives such as the Apache 2.0 editions of QuestDB, GreptimeDB and CrateDB can be self-run without a retention cap from the vendor.
Which alternatives are open source?
QuestDB, GreptimeDB, CrateDB and ClickHouse each describe an Apache 2.0 core, and Prometheus uses Apache 2. Grafana Mimir is AGPLv3. Apache IoTDB and Apache Druid are Apache Software Foundation projects. TDengine calls one edition fully open-source without naming the licence.
Is any alternative based in Europe?
CrateDB is the one that states a European origin, founded in Austria and offering a managed cluster on STACKIT.
Most others are US-based or state no country on the pages we read. Check the legal entity that would sign your contract, not only the product's origin.
Which alternative suits Prometheus-style monitoring?
Prometheus itself, VictoriaMetrics and Grafana Mimir all work in that world. VictoriaMetrics describes itself as a drop-in Prometheus replacement, while Mimir stores Prometheus data long term on object storage. Both are better matches for infrastructure metrics than for business records.
Can I try a paid alternative before committing?
Tiger Data gives a 30-day trial and InfluxDB Enterprise offers a 30-day free trial too. ClickHouse lists 30-day trial credit. GreptimeDB's pricing page mentions only a live demo, and QuestDB states a trial without its length.
—
Sources
Checked 7 November 2026
InfluxData pricing: Serverless meters ($0.0025 per MB, $0.012 per 100 queries, $0.002 per GB-hour, $0.09 per GB egress), 30-day retention on free tier, Core, Enterprise 30-day trial, Dedicated and Enterprise priced by contact, Amazon Timestream for InfluxDB.
InfluxDB 3 Core docs: Open source release, Parquet persistence, SQL and InfluxQL sections, Enterprise adds high availability and read replicas, Flux only under InfluxDB 2.
Founder of The Knowledge Engineers and of 2 Steps Ahead, an AI search visibility agency. Editorially responsible for every page on this site; research and drafting are AI-assisted.
Author and fact-checker: Sebastiaan Smits. This page was last reviewed on 7 November 2026; our review process is public.
