AIConfigurator
AIConfigurator workload (test_template_name is Aiconfigurator) runs the AIConfigurator predictor using the installed aiconfigurator Python package. It is a Standalone workload (no Slurm/Kubernetes/RunAI required).
Outputs
Each test run produces:
report.json: Predictor output (JSON dict of metrics and metadata)stdout.txt/stderr.txt: Predictor logsrun_simple_predictor.sh: Repro script containing the exact executed command (useful for debugging)
Usage Example
Test TOML example (Disaggregated mode):
Test TOML example (Aggregated/IFB mode):
Running
API Documentation
Command Arguments
AiconfiguratorCmdArgs
Command arguments for Aiconfigurator workload with nested agg/disagg configs.
Test Definition
AiconfiguratorTestDefinition
Test object for running Aiconfigurator predictor as a workload.
Command Generation Strategy (Standalone)
cloudai.workloads.aiconfig.standalone_command_gen_strategy.AiconfiguratorStandaloneCommandGenStrategy
Generate a standalone command that invokes the Aiconfigurator predictor and writes JSON output.
cleanup_job_artifacts
Best-effort cleanup hook run after the job has fully completed.
Report Generation Strategy
cloudai.workloads.aiconfig.report_generation_strategy.AiconfiguratorReportGenerationStrategy
Generate metrics from Aiconfigurator predictor outputs.