This is a major update: the FaaSr R package has been repurposed as a
local workflow testing and development environment. It
no longer requires cloud infrastructure or credentials to run FaaSr
workflows.
The core backend execution logic has been migrated to the FaaSr-Backend
repository, which handles production deployment across GitHub Actions,
AWS Lambda, OpenWhisk, Google Cloud, and SLURM.
Use this package to develop and validate FaaSr workflow JSON
configurations locally before deploying to any cloud FaaS platform via
the backend.
FaaSr 1.4.4
Support optional argument to S3 functions instead of global
.faasr
Workflow cycle detection function now covers the entire DAG
FaaSr 1.4.3
Support triggering of multiple parallel executions from an
action
Added faasr_rank() method to determine the rank of a parallel
invocation
Support passing of payloads through API invocation for GitHub
Actions
Improvement of faasr_s3_check() to support AWS S3 buckets
FaaSr 1.3.0
Support anonymous access to arrow/S3 buckets
Self-abort on a timeout while waiting for a lock
Improve credential management in RStudio
Add logging functions to faasr_client_tools.R
FaaSr 1.2.1
Fix bugs for local execution functions.
FaaSr 1.2.0
Print the UUID and logs folder to help users locate logs when an
action fails
Implement faasr_get_folder_list function to retrieve S3 folder
contents as a list
Return without action failure on
faasr_abort_on_multiple_invocations
Support local execution on client for development and debugging
Fixed bug that limited maximum memory size in AWS Lambda
Increased default size of /tmp storage in AWS Lambda
FaaSr 1.1.2
This version incorporates CRAN review feedback from initial
submission