Single-Node Clusters
Single-node clusters are used for running jobs on a single machine
Clusters
Clusters define hardware specifications for compute jobs run on Kaspian
Clusters are configured in the Settings > Clusters tab on the left-hand side. Click on the + symbol next to New Cluster. Kaspian currently supports two types of compute clusters. Single-node clusters are useful for running any Python jobs, including those involving ETL on small datasets or PyTorch deep learning models. Spark clusters are useful for running large-scale ETL jobs.