I work on a proprietary piece of software written in C# with SQLServer for storage. I evaluate possibilities of a full/partial replacement of storage with NoSQL Database (mainly for time series analytics) and was considering using Cassandra. I successfully connect and use Cassandra using provided C# driver but for analytics, I understand an efficient way would be using Spark. Therefore, I try to setup a config where I can run basic SQL queries (group by mainly) against a Spark/Cassandra cluster using Spark SQL and ODBC Driver (ODBC driver for Spark exists). Without any success for now.
Is anyone already used such a config ? Is it possible ? Else, could you point me to the right direction ?
Thanks in advance.