Currently R can import data from excel and csv file.
I doubt that it can read in data from SQL Server smoothly while SAS can handle almost all data source.
The standard R Import-Export documentation (PDF) describes how to access databases. In the Windows context, this amounts to creating an ODBC link, and then accessing it directly from R.
Many people (including me!) use R to access relational databases via SQL directly. As a local database, you can always install something like PostgreSQL and use R to write to it for later retrieval, or you can use something like the database package Filehash for such storage.
You may also want to take a look at the open source package KNIME. It has an excellent visual interface, can read from a database, and it integrates with R.
Hello Ram,
One possible way to use SQL server with R is via package RODBC. You should try it first. Then you can go for RDBI package.
I recommend R for using as data mining tool. There are thousands of packages. You should find out what people did in those packages and then you can write your own functions over others work. You dont have to reinvent the wheel, do you?