Reputation: 10099
Are the databases offered by cloud services such as Windows Azure SQL Database or AWS Big Data capable of distributed computing, in the sense that the query optimizer divides the work across servers which compute in parallel, similar to how MapReduce distributes computation across nodes?
I haven't found anything about any such query optimization in the Azure documentation, although PDW seems like it may do this.
Upvotes: 0
Views: 145
Reputation: 1177
AWS has EMR (Elastic Map-Reduce) which is Hadoop provisioned by AWS.
Azure has HDInsights which is Hortonwork's data platform (Hadoop) installed on Windows VMs.
Microsoft's PDW (parallel data warehouse) doesn't support map-reduce right as far as I know but they are working on it (http://www.zdnet.com/microsofts-polybase-mashes-up-sql-server-and-hadoop-7000007424/) - PDW is essentially a few SQL Server machines with a central management layer that allows partitioning and distribution of the data between the different nodes - it can and will break a query between the PDW nodes if the data resides on more than one but the parallelism is not map-reduce in nature.
Upvotes: 1