Insert Statement for 2 million rows

Question

Insert Statement for 2 million rows

PJ_SQL

SSCertifiable

Points: 6703
More actions
March 13, 2014 at 10:11 am

#150637

Hi!
I have multiple (individual) insert statement to insert around 2 million rows (using sqlcmd because the insert stament is around 8 thou lines).
It took about an hr to insert when I tried it on my machine which is physical server with 4 GB RAM.
When I try to run the same on my test server which is virtual machine it is still running,we have bumped up memory from 2 to 4 GB RAM.
Any suggestions?

Viewing 8 posts - 1 through 7 (of 7 total)

You must be logged in to reply to this topic. Login to reply

Jeff Moden SSC Guru Points: 1004340 More actions · Answer 1

pretea (3/13/2014)
Hi!
I have multiple (individual) insert statement to insert around 2 million rows (using sqlcmd because the insert stament is around 8 thou lines).
It took about an hr to insert when I tried it on my machine which is physical server with 4 GB RAM.
When I try to run the same on my test server which is virtual machine it is still running,we have bumped up memory from 2 to 4 GB RAM.
Any suggestions?

Why is the insert statement around 8 thousand lines?

--Jeff Moden

RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

Change is inevitable... Change for the better is not.

Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)

PJ_SQL SSCertifiable Points: 6703 More actions · Answer 2

It is how the vendor sent in the script?

Other thing I can try is to ask them to load data into text file and use BCP.

Jeff Moden SSC Guru Points: 1004340 More actions · Answer 3

pretea (3/13/2014)
It is how the vendor sent in the script?
Other thing I can try is to ask them to load data into text file and use BCP.

Heh... I can only imagine what that script must look like.

Yes, I agree. Load the text file into a staging table using BCP, validate the data, and then insert it into the final table.

--Jeff Moden

RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

Change is inevitable... Change for the better is not.

Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)

Eric M Russell SSC Guru Points: 125561 More actions · Answer 4

pretea (3/13/2014)
Hi!
I have multiple (individual) insert statement to insert around 2 million rows (using sqlcmd because the insert stament is around 8 thou lines).
It took about an hr to insert when I tried it on my machine which is physical server with 4 GB RAM.
When I try to run the same on my test server which is virtual machine it is still running,we have bumped up memory from 2 to 4 GB RAM.
Any suggestions?

Performance on a workstation PC, much less a VM running on your PC, will bear little resemblance to performance on a typical production server. That said, first option should be to get a delimited text version of the records and then BCP into database, like others suggested. You can use 'Export Data' wizard to copy records from original source datbase to text file.

The problem with a script is that all 2 million inserts are probably getting batched into one huge transaction. If all you have to work with is this single script, then use a tool like GSplit to chop it into seperate smaller batches. I'm guessing 20 .sql scripts with 100,000 inserts per batch should take it down to a manageable level. Also, you can use PowerShell script or SQLCMD command and a batch file to execute the scripts sequentially.

http://www.gdgsoft.com/gsplit/

http://technet.microsoft.com/en-us/library/ms170572.aspx

"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho

Jeff Moden SSC Guru Points: 1004340 More actions · Answer 5

Eric M Russell (3/13/2014)
pretea (3/13/2014)
Hi!
I have multiple (individual) insert statement to insert around 2 million rows (using sqlcmd because the insert stament is around 8 thou lines).
It took about an hr to insert when I tried it on my machine which is physical server with 4 GB RAM.
When I try to run the same on my test server which is virtual machine it is still running,we have bumped up memory from 2 to 4 GB RAM.
Any suggestions?
Performance on a workstation PC, much less a VM running on your PC, will bear little resemblance to performance on a typical production server. That said, first option should be to get a delimited text version of the records and then BCP into database, like others suggested. You can use 'Export Data' wizard to copy records from original source datbase to text file.
The problem with a script is that all 2 million inserts are probably getting batched into one huge transaction. If all you have to work with is this single script, then use a tool like GSplit to chop it into seperate smaller batches. I'm guessing 20 .sql scripts with 100,000 inserts per batch should take it down to a manageable level. Also, you can use PowerShell script or SQLCMD command and a batch file to execute the scripts sequentially.
http://www.gdgsoft.com/gsplit/
http://technet.microsoft.com/en-us/library/ms170572.aspx

BCP will actually do all of that.

--Jeff Moden

RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.

Change is inevitable... Change for the better is not.

Helpful Links:
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)

Eric M Russell SSC Guru Points: 125561 More actions · Answer 6

Jeff Moden (3/13/2014)
Eric M Russell (3/13/2014)
pretea (3/13/2014)
Hi!
I have multiple (individual) insert statement to insert around 2 million rows (using sqlcmd because the insert stament is around 8 thou lines).
It took about an hr to insert when I tried it on my machine which is physical server with 4 GB RAM.
When I try to run the same on my test server which is virtual machine it is still running,we have bumped up memory from 2 to 4 GB RAM.
Any suggestions?
Performance on a workstation PC, much less a VM running on your PC, will bear little resemblance to performance on a typical production server. That said, first option should be to get a delimited text version of the records and then BCP into database, like others suggested. You can use 'Export Data' wizard to copy records from original source datbase to text file.
The problem with a script is that all 2 million inserts are probably getting batched into one huge transaction. If all you have to work with is this single script, then use a tool like GSplit to chop it into seperate smaller batches. I'm guessing 20 .sql scripts with 100,000 inserts per batch should take it down to a manageable level. Also, you can use PowerShell script or SQLCMD command and a batch file to execute the scripts sequentially.
http://www.gdgsoft.com/gsplit/
http://technet.microsoft.com/en-us/library/ms170572.aspx
BCP will actually do all of that.

Yes, BCP is best option, unless all that's available is the insert script. We can not BCP a .sql insert script.

"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho

PJ_SQL SSCertifiable Points: 6703 More actions · Answer 7

Yes let's see what see, have asked them to get the text version of the records and use then we can use BCP if not use smaller batches.

Thanks for all the suggestions.