Overview
alchemiscale is described as a high-throughput alchemical free energy execution system for HPC, cloud, bare metal and Folding@Home. Its documentation covers use and deployment separately from the project site, and the Python package is published on the package index.
Deploying it means running a service that stores and scopes alchemical tasks and one or more compute services that execute them, so access control, resource accounting and the network boundary between the service and compute resources are operator decisions. Jobs sent to external or volunteer infrastructure carry their own data-egress and acceptable-use considerations, and the scientific validity of queued calculations still depends on the underlying protocols and convergence checks.
LICENSE states the MIT licence and the project is hosted by Open Free Energy. External simulation engines keep their own terms.
Key Features
- Task and transformation workflows distributed to HPC, cloud, bare-metal and volunteer compute resources.
- Separate documentation for client use and for service deployment.
Use Cases
- Queue a large set of alchemical transformations across available compute resources and collect the results centrally.
How to Use
Install the client package and connect it to an alchemiscale service, or follow the deployment documentation to run your own. Decide where task data may be sent, and keep protocol settings and convergence diagnostics with the results.
See the official documentation for client and deployment guidance: https://docs.alchemiscale.org/en/latest/
Sources and verification
This entry describes capabilities documented by the linked project sources. Catalog edits are not evidence of a successful run, compatibility test or scientific performance evaluation.
Read the editorial and verification policy