We are excited to announce that Replicated Tables, a new type of table distribution, are now generally available in Azure SQL Data Warehouse (SQL DW). SQL DW is a fully managed, flexible, and secure cloud data warehouse tuned for running complex queries fast and across petabytes of data.
The key to performance for large-scale data warehouses is how data is distributed across the system. When queries join across tables and data is distributed differently, data movement is required to complete the query. The same can be said when transforming data to load, enrich, and apply business rules. With Replicated Tables, the data is available on all compute nodes, hence data movement is eliminated, and queries run faster. In some cases, such as small dimension tables, choosing a Replicated Table versus a Round Robin table, can increase performance because data movement is reduced. As with all optimization techniques, performance gains may vary and should be tested.
Read about it in the Azure blog.
