Hi,
I have taken a look online and on forums, but I cannot seem to find information on the limitations (or lack thereof) of sharding on SQL.
https://azure.microsoft.com/en-us/documentation/articles/sql-database-elastic-scale-introduction/
Doing joins on a sharded cluster is something that is not normally architecturally possible.
MySQL NBD cluster, however, is able to perform JOIN operations across a partition due to the presence of distinct data nodes and SQL nodes.
https://www.mysql.com/products/cluster/scalability.html
Does Microsoft Azure SQL and Microsoft SQL server support this functionality as well? If it is possible, is the performance fast enough for production workloads? Are there any other functional limitations to horizontal scaling on Azure SQL?