DEV.co
Open-Source Observability · Basekick-Labs

arc

Arc is a single-binary analytical database written in Go that handles high-throughput data ingestion (19M+ records/sec), SQL queries (8M+ rows/sec), and automated data lifecycle management. It stores data as open Parquet files on S3/Azure without vendor lock-in, and requires no external dependencies or cluster infrastructure.

Source: GitHub — github.com/Basekick-Labs/arc
622
GitHub stars
38
Forks
Go
Primary language
AGPL-3.0
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
RepositoryBasekick-Labs/arc
OwnerBasekick-Labs
Primary languageGo
LicenseAGPL-3.0 — OSI-approved
Stars622
Forks38
Open issues58
Latest releasev26.06.3 (2026-07-02)
Last updated2026-07-03
Sourcehttps://github.com/Basekick-Labs/arc

What arc is

Arc is a columnar OLAP database built in Go that combines a high-performance ingestion pipeline, DuckDB query engine, automatic background compaction, retention policies, continuous queries, and MQTT support in a single statically-linked binary. It uses MessagePack and Arrow IPC wire formats alongside JSON, supports standard SQL with window functions and CTEs, and persists to Parquet on cloud object storage.

Quickstart

Get the arc source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/Basekick-Labs/arc.gitcd arc# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

High-throughput event analytics

Product analytics, clickstreams, user behavior tracking, and A/B testing where ingestion rates exceed 10M events/sec and multi-dimensional queries are required.

Observability at scale (metrics, logs, traces)

Distributed systems monitoring, time-series metrics, structured logs, and trace storage where air-gap capability, on-premise deployment, and cost predictability matter more than managed SaaS.

Edge and tactical deployments

Manufacturing telemetry, IoT sensor networks, MQTT data streams, battlefield/disconnected operations, and sovereign cloud environments where a single binary with zero external runtime dependencies is essential.

Implementation considerations

  • AGPL-3.0 license: any modifications to Arc or integration as a networked service may require source code release. Legal review is required before production deployment in proprietary environments.
  • Single binary simplifies deployment but requires explicit operational runbooks for backup/restore, compaction tuning, and retention policy lifecycle management.
  • Performance benchmarks (19M rec/sec ingestion, 8M+ rows/sec queries) are from controlled M3 Max environments; production performance varies with schema, compression, query complexity, and cloud storage latency.
  • Data is immutable-append-only (Parquet files on object storage). Supports soft deletes and compaction but not row-level UPDATE/DELETE; schema evolution requires planning.
  • Continuous query and retention policy scheduling are embedded; no external scheduler required, but operational monitoring and alerting are operator responsibility.

When to avoid it — and what to weigh

  • Need OLTP (row-oriented transactions) — Arc is columnar and optimized for analytical workloads. If your use case is frequent single-row inserts, updates, deletes, or strict ACID transactions on mutable data, consider PostgreSQL or MySQL.
  • Require proprietary commercial support SLA — Arc is AGPL-3.0 licensed. Commercialization or closed-source derivatives require vendor agreement. If you need guaranteed enterprise support contracts, verify with Basekick Labs first.
  • Cannot accept copyleft license obligations — AGPL-3.0 requires source disclosure for networked software modifications. If your compliance policy forbids copyleft or you cannot share modifications, this license is a blocker.
  • Prefer managed, hands-off analytics — Arc requires operational ownership: schema design, data ingestion pipelines, retention policies, and continuous query management. If you want a fully managed analytics platform, consider Snowflake or Databricks.

License & commercial use

Arc is licensed under AGPL-3.0 (GNU Affero General Public License v3.0). This is a strong copyleft license requiring that any derivative work or networked modification disclose source code to users. Statically linking Arc into a proprietary application or running an unmodified Arc binary as-is is permitted; any modifications to Arc's source or running Arc as a service component require source disclosure.

Arc's AGPL-3.0 license does not explicitly prohibit commercial use but imposes source code disclosure requirements for networked derivatives. If Arc is deployed unmodified as a standalone analytical database for internal use, commercial deployment is permissible. If Arc is modified, embedded in a larger service, or offered as a SaaS product, source code disclosure is legally required. Contact Basekick Labs to discuss commercial licensing (dual-license or proprietary exceptions) if copyleft compliance is not feasible.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationAdequate
License clarityNeeds review
Deployment complexityLow
DEV.co fitGood
Assessment confidenceMedium
Security considerations

No security audit, pen test, or CVE history is visible. AGPL-3.0 source is available for review. Authentication and data encryption mechanisms are not described in provided data. Object storage security depends on cloud provider (S3 IAM, Azure RBAC). Air-gap deployment eliminates cloud-based attack surface but requires secure local infrastructure. No formal security policy or responsible disclosure process stated. Requires security review before handling sensitive data.

Alternatives to consider

ClickHouse

Mature open-source columnar database (Apache 2.0) with distributed cluster support, enterprise backing (Yandex), and stronger documentation. Requires cluster management and external coordination (ZooKeeper). More complex operational overhead but broader adoption and commercial support options.

Druid

AGPL-3.0 time-series OLAP database optimized for real-time analytics. Requires distributed infrastructure (ZooKeeper, deep storage). Stronger ecosystem for observability and monitoring use cases but higher operational complexity than Arc's single binary.

DuckDB (standalone + custom ingestion)

Permissive MIT license, embedded or server mode, excellent SQL support. Arc wraps DuckDB as its engine. If you do not need Arc's opinionated ingestion pipeline, compaction, or retention policies, DuckDB alone reduces licensing concern but shifts operational burden to custom code.

Software development agency

Build on arc with DEV.co software developers

Review the live demo at basekick.net/demos, assess AGPL-3.0 license fit with your legal team, and test ingestion/query performance on your data schema. Join the Discord community for deployment guidance.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

arc FAQ

Can Arc be used in a proprietary product without source disclosure?
Only if Arc is deployed unmodified as a standalone service. If Arc is modified, embedded in proprietary code, or run as part of a networked SaaS platform, AGPL-3.0 requires source code disclosure. Dual-licensing or commercial exceptions may be available from Basekick Labs.
Does Arc replace a data warehouse like Snowflake?
Arc is a database, not a fully managed data warehouse. It handles ingestion, storage, and SQL queries but does not provide Snowflake's managed infrastructure, automatic scaling, or integrated BI tools. Arc is better suited for on-premises, edge, or cost-sensitive deployments where operational ownership is acceptable.
How does Arc handle schema changes or data updates?
Arc is append-only; it does not support traditional UPDATE/DELETE at the row level. Compaction and retention policies manage data lifecycle. Schema changes (adding columns) are supported but not documented in the excerpt. Hard deletes require manual intervention or table recreation.
What happens if Arc crashes or loses data?
Arc persists to Parquet on object storage (S3/Azure), so data survives process crashes. Backup and restore procedures are not detailed in the provided excerpt. Operational procedures for high availability (replication, disaster recovery) are not evident; requires further documentation review.

Work with a software development agency

DEV.co is a software development agency delivering custom software development services to companies building on open source. Our software developers and web developers design, integrate, and ship production systems — spanning web development, APIs, AI, data, and cloud. If arc is part of your open-source observability roadmap, our team can implement, customize, migrate, and maintain it.

Evaluate Arc for Your Analytics Workload

Review the live demo at basekick.net/demos, assess AGPL-3.0 license fit with your legal team, and test ingestion/query performance on your data schema. Join the Discord community for deployment guidance.