LALInference Pipelines
Asimov provides full support for the LALInference pipeline.
While LALInference has been largely superseded by newer sampling techniques it can still be helpful to be able to run jobs using it both for carrying-out cross checks, and for replicating older analyses.
Review status
Note
The current integration with LALInference is fully reviewed and is suitable for use in collaboration analyses.
Examples
LALInference with Markov Chain Monte Carlo
- Prod0:
pipeline: lalinference
approximant: IMRPhenomXPHM
nparallel: 25
engine: lalinferencemcmc
status: ready
LALInference with Nested sampling
- Prod0:
pipeline: lalinference
approximant: IMRPhenomXPHM
nparallel: 25
engine: lalinferencenest
status: ready
LALInference interface
The LALInference interface can be used to submit jobs to condor clusters which use lalinference as the inference engine.
LALInference jobs must be specified by an appropriately formatted ini
file.
Status messages
Events run by the LALInference pipeline can have the following status values:
wait
In this state the pipeline will ignore the production
ready
In this state asimov will attempt to submit the job to the condor scheduler
running
Applied after the job is submitted to the cluster
stuck
Applied when the job is held or an error is detected in the pipeline’s execution
finished
Applied when normal termination of the pipeline is detected.
Run data
The following values will be added to the production meta data by asimov as a production is running
user
The accounting user who submitted the event.
job id
The JobID for this event on the condor cluster.