analysis_scheduling

Specify the scheduling of concurrent analyses when Dakota is run in parallel

Topics

concurrency_and_parallelism

Specification

  • Alias: None

  • Arguments: None

  • Default: automatic (see discussion)

Child Keywords:

Required/Optional

Description of Group

Dakota Keyword

Dakota Keyword Description

Required (Choose One)

Scheduling Mode

master

Specify a dedicated master partition for parallel analysis scheduling

peer

Specify a peer partition for parallel analysis scheduling

Description

When Dakota is run in parallel, the partition type for the analysis servers is determined automatically. If this setting is undesirable, it may be overridden by the user using the analysis_scheduling keyword.