Thanks Dave.

I am working with win server 2003 with SQL Server 2005. I wanted to try back ups(just for working knowledge sake) and due to limited resources i wanted to try this using 2 instance of SQL rather than 2 different servers.

I had database on default instance and i wanted to restore it on named instance 1. Since i want to maintain the same name of the database i tried to create a database with same name and i got an error which i was expecting because i already had a database from named instance on the PRIMARY file group. Hence while creating the database on the instance 1 i created a new file group and changed it to default. However i had to create a new file called mydatabase_1.mdf and only then the process was sucessful. However while i was creating the datbase it still had the mydatabase.mdf and mydatabase_log.ldf, the third file i created was mydatabase_1.

I am confused as to which will would have been present on which file group.

Will you please explain what is the relationship between filegroup and table is it a One-to-One or One-to-many. Can we write multiple tables into single filegroup which has multiple files?? If we can how to retrieve multiple tables data from a single file which is in a filegroup?? Is my question is ODD?

for example i have a database which has three large tables A ,B and C. i want to partition month by month data for june,july,aug,sep for each table A,B and C . To achieve this do i need to create 12 files and 12 filegroups for the 3 tables?? Please Explain.

Thanks
Siri

Thanks Dave.

I want to add file group dynamically.

Ex.
When date change i execute script for create file group as per date.

Declare @FileName varchar(50)
Set @FileName=convert(varchar,getdate(),112)
print @FileName
ALTER DATABASE VehicleTrackingSystem ADD FILEGROUP [@FileName]

Above script create file group with name @FileName which is wrong.

Please look in this problem.

Regards,
Parthesh

At the time of creation you can specify the filegroup as below:

CREATE TABLE tableName (columns specification…) ON filegroup

An existing table (heap) can be moved to another filegroup by creating the clustered index on new filegroup as below:

CREATE CLUSTERED INDEX indexName ON tableName (column…) ON filegroup

Now if the clustered index is not required then it can be dropped.

If table already have a clustered index then create the clustered index with DROP_EXISTING option.

Regards,
Pinal Dave

Asking question is not funny and you asked a very good question. You are comparing two methods (partitioning and storing a table on separate filegroup) to split the table data over multiple disks files.
If a filegroup has multiple files and you store a table on this filegroup, table’s data would be splitted over all files. This is a method to improve the I/O performance of a table. But data splitting over multiple files is not visible to user and we can not use this method to work upon a section on data.
Partitioning is another way to split the data over multiple filegroup (and internally over multiple files). But in this method we know which record is stored in which filegroup. The partitions of data are visible to user and we can perform few tasks like SELECT, index defragmentation, etc. on a single partition also.

Regards,
Pinal Dave

Can you please help?

Regards

Sutanu

I’m very new to this technology. In the above you told about
Create Multiple Filegroup For Single Database..i have small query how to add table to this new filegroup…

I am New to MS sql, But I find interesting in Database servers, As I would like to learn the Concepts of Administration the SQL server as will as how efficient can I make the server work.

Can you Please Help me in it

regards
SreeMon