We created an Virtual Machine instances that uses MapR-FS as a datastore (technically all VM data including image is stored on the MapR-FS)
I mounted an NFS directory on Microsoft Server. Using that directory, I directly dump .mdf files from SQL Server.
Can I even process or query on these files using drill, spark, or mapreduce?
Or it is exclusive only on SQL Server and SQL server would be the only way to read and process those files.
Also If these files are unreadable using the options above, Is it more efficient to run sqoop given they are in the same network(LAN) ?