Cluster Configurations - 2024.1 English

Vivado Design Suite User Guide: Implementation (UG904)

Document ID
UG904
Release Date
2024-06-05
Version
2024.1 English

Compute Clusters are groups of machines configured through third party tools that accept jobs, schedule them, and efficiently allocate the compute resources. Common compute clusters include LSF, SGE and SLURM. To add custom compute clusters to Vivado, you can click the plus tool bar button shown in figure def and provide a name for the cluster configuration. You then need to specify the command necessary to submit a job to the cluster, cancel a job on the cluster, and the cluster type. Vivado natively supports LSF, SGE, and SLURM. For any other cluster, you can choose CUSTOM in the combo box. For CUSTOM cluster, you should provide path of the Tcl file, which contains logic to fetch Job ID and return Job ID value to the proc. This proc name should be used to populate the field Job ID Proc. Job ID Tcl and Job ID Proc can be left empty for natively supported clusters, that is, LSF, SGE, and SLURM. The configuration can be tested by pressing the test configuration button.

Figure 1. Cluster Configurations Settings Dialog Box