GreptimeDB vs. InfluxDB

Line protocol wrote it.
TSM compacted it.

InfluxDB is a time-series database developed by InfluxData for monitoring, IoT, and analytics. It evolved through multiple versions: v1/v2 use the TSM storage engine with InfluxQL and Flux query languages, while v3 adopts Apache Parquet and Arrow. v3 Core is released under MIT/Apache 2.0, but v2.x remains under BSL 1.1. The open-source v3 Core build is single-node and optimized for recent data, while clustering, high availability, and unlimited historical retention live in the commercial Enterprise and Cloud tiers. The version fragmentation — different storage formats, query languages, and licensing across major versions — is a key consideration for adopters. Click here to read the full performance comparison report between GreptimeDB and InfluxDB.

CHALLENGER

InfluxDB

Version-fragmented · TSM (v1/v2), Parquet (v3) · Flux deprecated in v3

GREPTIMEDB

GreptimeDB

Apache 2.0 · open Parquet storage · InfluxDB Line Protocol compatible

Feature comparison

Feature/Aspect	GreptimeDB	InfluxDB
Data Model	Metrics, Logs & Traces in one database	Time-Series Database
Value Model	Multi-Value (supports complex data structures)	Multi-Value (tags and fields)
Multi-model Support	Metrics, Logs & Traces in one database	Metrics only (requires separate systems for logs/traces)
Ingestion Protocols	SQL gRPC InfluxDB Line Protocol Prometheus Remote Storage OpenTelemetry Loki Push API Elasticsearch Bulk API HTTP API	InfluxDB Line Protocol HTTP API Telegraf plugins
Query Languages	SQL & PromQL (dual interface)	InfluxQL & Flux (v1/v2) SQL & InfluxQL (v3)
Data Retention	Flexible TTL policies with tiered storage	Retention policies with automatic deletion
Continuous Aggregation	Built-in SQL aggregation, Pipeline ETL engine & Flow streaming computation	Continuous queries (v1/v2) Tasks and checks (v3)
Use Cases	Unified observability, real-time analytics, IoT monitoring, edge computing	IoT monitoring, DevOps metrics, application monitoring
Architecture	Cloud-native distributed with compute-storage disaggregation	Single binary (v1/v2) Cloud-native microservices (v3)
Storage Format	Apache Parquet (columnar, compressed)	TSM engine (v1/v2) Apache Parquet (v3)
Storage Scalability	Object storage integration with unlimited capacity	Local storage (v1/v2) Object storage support (v3)
High Availability	Native clustering with automatic failover	Enterprise clustering (v1/v2) Native HA (v3)
License	Apache 2.0	MIT/Apache 2.0 (v3 Core) BSL 1.1 (v2.x) Commercial (Enterprise/Cloud)
Open-source Scope	Full-featured open source (Apache 2.0) — same engine from single node to cluster	v3 Core is single-node, recent-data; clustering, HA, and historical query are in Enterprise/Cloud
Deployment Options	Single-node, cluster, Kubernetes-native, edge-to-cloud with unified API	Single binary, containerized, cloud-managed
Version Compatibility	Consistent API from edge to cloud with unified control plane	Breaking changes between major versions
Written Language	Rust (memory safety, performance)	Go (v1/v2) Rust (v3)

Redirect write requests

Docs

Point InfluxDB v1/v2 write APIs to the GreptimeDB endpoint - same line protocol, only URL, credentials, and database name change.

30 min

Dual write and verify

Send new writes to both systems, compare row counts and key aggregation results, then shift read traffic step by step.

1-2 days

Backfill historical data

Export history as line protocol and batch import through the same write API while live ingestion continues.

1-3 days

Migrate InfluxQL queries and decommission

Rewrite InfluxQL as GreptimeDB SQL - range queries cover the window aggregations, and AI assistants handle most of the rewrite. After validation, stop writing to InfluxDB.

1 Week