{question}
Is it advantageous to have a smaller number of leaf nodes with augmented resources compared to the others?
{question}
{answer}
The short answer here is NO.
SingleStore engine stores data on database partitions, and the engine is designed to spread the workload equally across all nodes ensuring that the workload is evenly distributed across all nodes. This means that even if one node has more CPUs or memory, it will still store the same number of partitions as the other nodes, rather than using its full potential.
When it comes to query processing, every query is executed on an aggregator node, which instructs the leaf nodes and collects partial results from the leaf nodes to aggregate the final result of the query. Even if one node is faster than other nodes and provides the results faster, the aggregator needs to get data from other nodes as well to present you with a consistent result of the query.
Reference
Interaction of Cluster Components
{answer}