ArmoniK is a high throughput compute grid project using Kubernetes. The project provides a reference architecture that can be used to build and adapt a modern high throughput compute solution on-premise or using Cloud services, allowing users to submit high volumes of short and long-running tasks and scaling environments dynamically.
Warning: This project is an Open Source (Apache 2.0 License).
ArmoniK should be used when the following criteria are meet:
- A high task throughput is required (from 250 to 10,000+ tasks per second).
- The tasks are loosely coupled.
- Variable workloads (tasks with heterogeneous execution times) are expected and the solution needs to dynamically scale with the load.
ArmoniK might not be the best choice if :
- The required task throughput is below 250 tasks per second.
- The tasks are tightly coupled, or use MPI.
- The tasks use third party licensed software.
The current version of ArmoniK uses the tags listed in armonik-versions.txt where:
core
is the ArmoniK Core tag used for container images of Control plane, Polling agent and Metrics exporter.worker
is the tag used for the container image of the workersadmin-gui
is the tag used for the container images of ArmoniK AdminGUI (admin-api and admin-app)samples
is the tag for ArmoniK Samples
All instructions to build, deploy and test ArmoniK software on Linux are described in ArmoniK deployment
Please Clone the repository Armonik.Samples into the [Root_Armonik_folder]/Source/
git clone https://github.com/aneoconsulting/ArmoniK.Samples.git
Instructions to run ArmoniK.Samples are described in Run Samples
This project was funded by AWS and started with their HTCGrid project.
Please direct enquiries about ArmoniK to the public mailing list [email protected].
See also Issues of ArmoniK project.
To report a bug or request a feature, please use and follow the instructions in one of the issue templates. Don't forget to include the version of ArmoniK you are using.