Skip to main content
OpenSourceAlternatives

Bacalhau

Open source · OSI-approvedStatus: unknownNew

Run container jobs where your data already lives

Compute-over-data orchestration for distributed workloads Bacalhau is a distributed compute orchestration framework from the bacalhau-project community that brings compute to the data instead of copying large datasets to a central cluster. You submit jobs through a CLI or YAML job spec, an orchestrator schedules tasks onto compute nodes with local data access, and results are published to configured storage. It suits log processing at the source, edge and IoT fleets, and distributed analytics where transfer cost, latency, or data-sovereignty rules make centralization painful.[1][2][3]

Quick verdict

Run container jobs where your data already lives Bacalhau is an Apache-2.0 open-source distributed compute orchestrator that runs Docker and WebAssembly jobs where the data already lives, reducing large data transfers. A single self-contained binary acts as client, orchestrator, and compute node, with storage integrations for S3, HTTP, IPFS, and local volumes. Commercial support is offered separately by Expanso.[1][2][3]

What Bacalhau can do

  • Docker and WASM execution engines Runs existing Docker container images and WebAssembly images as tasks without requiring a rewrite of current workflows.[1][2]

  • Single-binary client, orchestrator, node A single self-contained binary serves as client, orchestrator, and compute node, so a network can start small and add nodes as demand grows.[1][2]

  • Pluggable storage and publishers Reads inputs from S3, HTTP and HTTPS, IPFS, and local storage, and publishes results to local volumes, S3, or other configured backends.[1][2]

  • CLI and YAML job submission Jobs can be submitted imperatively with bacalhau docker run arguments or declaratively with a YAML job specification.[1][2]

  • Multiple job types Supports batch, ops, daemon, and service job types for different workflow requirements.[1]

  • Data-locality-aware scheduling The orchestrator schedules tasks based on resource availability, data location, and job requirements, advertising CPU, memory, GPU, and storage capacity per node.[1]

  • Partition-tolerant event ledger Nodes keep operating during network outages using local event ledgers and reconcile missed events when partitions heal.[1]

  • Auth, queueing, and web UI Recent releases added user authentication and authorization, TLS support, job queueing with prioritization, and an embedded web UI.[1]

Who will benefit most

  • Distributed log and dataset processing Teams processing logs and datasets at the source across regions, clouds, and edge devices benefit from parallel jobs that run where data was created.[1][2]

  • Edge fleets and sovereignty boundaries Edge and IoT fleets with transfer-cost, latency, or sovereignty constraints benefit because sensitive data can be processed within defined security perimeters.[1][2]

  • Vetted computation on protected data Organizations wanting cross-organizational computation on protected datasets without exposing raw data can allow vetted computations under a code-based permission model.[1][2]

Limitations to weigh

  • Not a full device or identity platform Bacalhau orchestrates Docker and WASM workloads but does not replace device management, identity, or storage systems, so those must be operated separately.[1]

  • Independent user reports unavailable No suitable original Reddit or G2 user reports were found during bounded research, so reported day-to-day experience beyond official documentation remains a gap.[1]

  • Self-operated cluster overhead Running a distributed network still requires provisioning orchestrator and compute machines with Docker installed, which is operational work compared with using a fully managed service.[1][2]

Compatibility and deployment

From single node to global cluster Bacalhau runs on a range of architectures from large VMs and GPUs to low-power edge and IoT devices, in single-node, regional-cluster, or global-cluster layouts.[1][2]

Apache-2.0 with Expanso support The software is Apache-2.0 licensed open source and self-hostable; enterprise support options with flexible licensing models are offered through Expanso.[1][2]

Questions worth asking

Is Bacalhau free to use?

Open source, paid support optional Yes. Bacalhau is Apache-2.0 licensed open source, so the software itself can be used, modified, and distributed under that license. Commercial support with flexible licensing models is offered separately through Expanso for teams that want it.[1][2]

Can Bacalhau be self-hosted?

Self-hostable from laptop to global fleet Yes. Bacalhau is designed to be self-hosted from a single node for development up to regional or global clusters, installed from a single binary with Docker on compute nodes.[1][2]

Does Bacalhau require coding or DevOps skills?

Container skills expected, no rewrite needed Jobs are submitted through a CLI or YAML job specs and compute nodes need Docker installed, so basic container and infrastructure skills are expected. Existing Docker images can be reused without major rewrites, which lowers onboarding effort.[1][2]

What is the monthly cost of running Bacalhau?

No public license price; infra-dependent There is no published per-month Bacalhau license fee found in official sources; costs depend on the infrastructure you run it on and any optional Expanso commercial support, which has no public price listed.[1][2]

Sources and editorial review

Source research. Reviewed on October 7, 2026. Written by Sanjib Sah. AI assisted; reviewed by a human editor.

We have not run or tested this product.

  1. Bacalhau: Distributed Compute Over Data

    official · Observed October 7, 2026

    Official project homepage describing compute-over-data positioning, Docker and WASM support, efficiency, security and governance, and commercial support through Expanso.

  2. bacalhau-project/bacalhau repository

    official · Observed October 7, 2026

    Repository README with project description, key features, job types, storage integrations, use cases, Apache-2.0 license, and install instructions. Observed stars around 872.

  3. Bacalhau architecture documentation

    official · Observed October 7, 2026

    Architecture guide covering orchestrator and compute node roles, job submission modes, modular execution and storage interfaces, data-centric scheduling, resilience, deployment models, and security.

  4. Edge orchestration buyer guide

    official · Observed October 7, 2026

    Technical buyer guide stating Bacalhau is an Apache-2.0 project for orchestrating Docker and WASM workloads and does not replace device management, identity, or storage systems.

  5. Bacalhau releases

    official · Observed October 7, 2026

    Release history showing v1.x features including authentication and authorization, TLS, queueing, embedded web UI, and enterprise support options.

Compared with these products

No explicit proprietary baseline is recorded.

Similar projects

Related projects with shared categories, technologies, or licensing.

Golem

GolemOpen

Run batch tasks on shared provider machines, or earn by sharing yours

Stars3
Commitjust now
LicenseGPL-3.0

Golem is a peer-to-peer network where requestors run batch tasks on machines shared by providers, paying per use in the GLM token. Tasks execute inside provider-hosted environments such as VM images, as a single task or as many parallel tasks spread across providers.

Coolify

CoolifyOpen

Deploy applications to your own servers with one click

Stars48.9K
Commit8h ago
LicenseApache-2.0

Coolify is a deployment platform that runs on servers you already pay for. It builds from a Git repository, provisions databases, issues certificates, and handles rollbacks, without the per-instance pricing of a hosted platform.

Qdrant

QdrantAIOpen

Open-source vector database for AI search at any scale

Stars35K
Commitjust now
LicenseApache-2.0

Qdrant is an Apache-2.0 open-source vector database written in Rust that serves filtered similarity search over embedding collections for AI retrieval.

ntfy

ntfyOpen

Self-hosted push notifications over simple HTTP PUT/POST

Stars34.8K
Commitjust now
LicenseApache-2.0

ntfy is a self-hosted push-notification service that sends alerts to phones and desktops with a simple HTTP PUT or POST to a topic, plus open-source Android, iOS, web, and CLI apps.

Kestra

KestraOpen

Open-source orchestration platform for scheduled and event-driven workflows as code

Stars29.5K
Commitjust now
LicenseApache-2.0

Kestra is an open-source, event-driven orchestration platform for building scheduled and event-driven workflows as declarative YAML flows, authored in code or a live-updating user interface.

Woodpecker CI

Woodpecker CIOpen

Self-hosted open-source CI/CD with container-based pipelines

Stars7,996
Commitjust now
LicenseApache-2.0

Woodpecker CI is a self-hosted, open-source CI/CD engine that runs YAML pipelines from the repository in containers on Docker, Kubernetes, or local backends.