█████╗ ██╗ ██╗ ██████╗ ███████╗██╗ ██████╗ ██╗ ██╗
██╔══██╗██║ ██╔╝██╔═══██╗██╔════╝██║ ██╔═══██╗██║ ██║
███████║█████╔╝ ██║ ██║█████╗ ██║ ██║ ██║██║ █╗ ██║
██╔══██║██╔═██╗ ██║ ██║██╔══╝ ██║ ██║ ██║██║███╗██║
██║ ██║██║ ██╗╚██████╔╝██║ ███████╗╚██████╔╝╚███╔███╔╝
╚═╝ ╚═╝╚═╝ ╚═╝ ╚═════╝ ╚═╝ ╚══════╝ ╚═════╝ ╚══╝╚══╝
AkôFlow is an open-source engine for orchestrating and executing container-based scientific workflows in the computing continuum. It was originally developed within the e-Science Research Group at the Institute of Computing, Fluminense Federal University (UFF).
Although initially focused on Kubernetes-based workloads, AkôFlow has evolved to support general containerized execution across multiple infrastructures.
-
To learn more about AkôFlow, please visit our project page: https://akoflow.com/
-
To see our documentation, please visit: https://uffescience.github.io/akoflow/
Download the current installer for macOS, Windows or Linux from AkôFlow Releases.
AkôFlow Desktop requires Docker Desktop on macOS and Windows, or Docker Engine with the Compose v2 plugin on Linux. At startup it verifies the requirements, explains anything that is missing, and starts the version-matched AkôFlow daemon and BuildKit containers automatically. Docker Desktop for Windows must use Linux containers.
Docker selects an available API port bound only to 127.0.0.1; no fixed host
port is reserved. BuildKit remains entirely inside the Compose network.
The packaged desktop application checks the releases in this repository for a new version. Updates are downloaded only after confirmation and installed when the application restarts. The desktop application, daemon image and BuildKit image use the same release version.
When the updated application starts, it pulls and health-checks the matching containers. Persistent Docker volumes are retained. If the new runtime cannot become healthy, AkôFlow restores the last healthy container version and reports the failed update instead of deleting data.
Release maintainers should configure MACOS_CSC_LINK,
MACOS_CSC_KEY_PASSWORD, MACOS_APPLE_ID,
MACOS_APP_SPECIFIC_PASSWORD, MACOS_TEAM_ID, WINDOWS_CSC_LINK and
WINDOWS_CSC_KEY_PASSWORD as GitHub Actions secrets. These credentials sign
and notarize the installers; they are never included in the application.
Versioned images are published to the GitHub Container Registry for
linux/amd64 and linux/arm64:
ghcr.io/uffescience/akoflow-daemonghcr.io/uffescience/akoflow-buildkit
The production Compose bundle and configuration instructions are available in
releases/.
New versions are released automatically when a semantic-version tag such as
v1.2.3 is pushed to this repository. Each release publishes the desktop
installers and versioned Docker images for the AkôFlow daemon and BuildKit.
See all releases: https://github.com/UFFeScience/akoflow/releases
- D.Sc. Daniel de Oliveira — Research Advisor
- Wesley Ferreira - @ovvesley — Maintainer - IC/UFF
- Liliane Kunstmann - COPPE/UFRJ
- Debora Pina - COPPE/UFRJ
- Raphael Garcia — IC/UFF
- Yuri Frota — IC/UFF
- Marcos Bedo — IC/UFF
- Aline Paes — IC/UFF
- Luan Teylo — INRIA/Université de Bordeaux
-
Ferreira, W., Kunstmann, L., Paes, A., Bedo, M., & de Oliveira, D. (2024, October).
AkôFlow: um Middleware para execução de Workflows científicos em múltiplos ambientes conteinerizados. In 39th Simpósio Brasileiro de Banco de Dados (SBBD) (pp. 27-39). SBC. (DOI:10.5753/sbbd.2024.241126.) -
Ferreira, W., Kunstmann, L., Garcia R., Bedo, M., & de Oliveira, D. (2025, October). Plug and Flow: Execução de Workflows Científicos em Contêineres com o Middleware
AkôFlow. In 40th Simpósio Brasileiro de Banco de Dados (SBBD). (Paper just accepted)
AkôFlow originated as a final undergraduate project and has since expanded with broader contributions and integrations. It continues to serve both academic and industrial workflow execution scenarios.