What is the default distribution used for a table in Synapse Analytics?

In Synapse Analytics, the default distribution method for a table is indeed Round-Robin. This distribution method assigns data to distribution blocks across the compute nodes in a balanced manner without any specific logic regarding data values. The primary benefit of Round-Robin distribution is its simplicity and ease of use, as it evenly spreads the data across all nodes, which can help in optimizing performance for certain queries.

Using Round-Robin as the default means that if a user does not specify a distribution strategy while creating a table, the system will automatically distribute the rows of the table across the distributions in a sequential manner. This can be particularly advantageous when the data does not have a natural key to use for more targeted distribution methods, such as HASH, or when replicating the entire table is not practical or necessary.

In situations where data is largely uniform, Round-Robin can offer optimal performance, making it a versatile choice for a default distribution strategy in Synapse Analytics.