![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hi: Any help with this would be appreciated. Thanks in advance. Environment: SQL 2000 SP3 all up to date on Win2K Server SP4 all up to date. Issue: Trying to import a very large text file (12GB) into a db table where the data was originally from. The data was exported based on a selective query (with a WHERE clause), then the table was truncated, and I am in the process of reimporting the data. The idea was to delete a bunch of rows as fast as possible. The export of the text file took about an hour, the import is taking six hours and still running. When I looked at the performance counters, it seemed to be writing and reading to the drive that holds the tempdb. I assumed that its writing the data to the tempdb db for the single shot insert - mind you I did not optimize the package by tweaking the switch in the package properties pages. I am now running out of time, I have to have the data in by ...hmmm.. half an hour ago. Would it be worth the while to wait for it or restart the whole thing with optimized package options for this type of import (which I have no idea at this point how to do, BOL maybe)? Thank you for your help. If you could CC me zebdi (AT) hotmail (DOT) com A.K. |
#3
| ||||||||
| ||||||||
|
|
Which Task? Transform data task. |
|
Why not use TSQL's BULK INSERT command No idea wether this would be faster than DTS... |
|
Does the destination table have 1. Indexes Yes, clustered only. 2. Triggers None. Is the destination DB in at most BULK LOGGED mode? the db is in Simple mode, I'm guessing it's not bulk |
|
re you doing this in batches BULK INSERT (BatchSize =) Nope, straight usage of the DTS package (db connection, |
|
What about bcp? I thought this is the same engine used by DTS?! |
|
Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.konesans.com - Consultancy from the people who know |
|
"A.K." <anonymous (AT) discussions (DOT) microsoft.com> wrote in message news:176a01c46f4b$97257020$a601280a (AT) phx (DOT) gbl... Hi: Any help with this would be appreciated. Thanks in advance. Environment: SQL 2000 SP3 all up to date on Win2K Server SP4 all up to date. Issue: Trying to import a very large text file (12GB) into a db table where the data was originally from. The data was exported based on a selective query (with a WHERE clause), then the table was truncated, and I am in the process of reimporting the data. The idea was to delete a bunch of rows as fast as possible. The export of the text file took about an hour, the import is taking six hours and still running. When I looked at the performance counters, it seemed to be writing and reading to the drive that holds the tempdb. I assumed that its writing the data to the tempdb db for the single shot insert - mind you I did not optimize the package by tweaking the switch in the package properties pages. I am now running out of time, I have to have the data in by ...hmmm.. half an hour ago. Would it be worth the while to wait for it or restart the whole thing with optimized package options for this type of import (which I have no idea at this point how to do, BOL maybe)? Thank you for your help. If you could CC me zebdi (AT) hotmail (DOT) com A.K. . |
#4
| |||
| |||
|
|
Thank you for the response: Which Task? Transform data task. Why not use TSQL's BULK INSERT command No idea wether this would be faster than DTS... Does the destination table have 1. Indexes Yes, clustered only. 2. Triggers None. Is the destination DB in at most BULK LOGGED mode? the db is in Simple mode, I'm guessing it's not bulk logged. re you doing this in batches BULK INSERT (BatchSize =) Nope, straight usage of the DTS package (db connection, txt file connection and Transform Data Task in between) What about bcp? I thought this is the same engine used by DTS?! Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.konesans.com - Consultancy from the people who know Thanks again Allan. "A.K." <anonymous (AT) discussions (DOT) microsoft.com> wrote in message news:176a01c46f4b$97257020$a601280a (AT) phx (DOT) gbl... Hi: Any help with this would be appreciated. Thanks in advance. Environment: SQL 2000 SP3 all up to date on Win2K Server SP4 all up to date. Issue: Trying to import a very large text file (12GB) into a db table where the data was originally from. The data was exported based on a selective query (with a WHERE clause), then the table was truncated, and I am in the process of reimporting the data. The idea was to delete a bunch of rows as fast as possible. The export of the text file took about an hour, the import is taking six hours and still running. When I looked at the performance counters, it seemed to be writing and reading to the drive that holds the tempdb. I assumed that its writing the data to the tempdb db for the single shot insert - mind you I did not optimize the package by tweaking the switch in the package properties pages. I am now running out of time, I have to have the data in by ...hmmm.. half an hour ago. Would it be worth the while to wait for it or restart the whole thing with optimized package options for this type of import (which I have no idea at this point how to do, BOL maybe)? Thank you for your help. If you could CC me zebdi (AT) hotmail (DOT) com A.K. . |
#5
| |||
| |||
|
|
-----Original Message----- TSQL's BULK INSERT and the BULK INSERT TASK are the same. mk:@MSITStore:C:\Program%20Files\Microsoft%20SQL% 20Server\80\Tools\Books\dts sql.chm::/dts_elemtsk2_5m0b.htm The transform data task with table lock on uses IRowsetFastLoad(BULK COPY rowsets) which is the same as bcp. mk:@MSITStore:C:\Program%20Files\Microsoft%20SQL% 20Server\80\Tools\Books\adm insql.chm::/ad_impt_bcp_8wc2.htm I personally would use BULK INSERT and I would use batches. Otherwise the data is going to be inserted in one big lump and that lump must be stored somewhere first. -- -- Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.konesans.com - Consultancy from the people who know "A.K." <anonymous (AT) discussions (DOT) microsoft.com> wrote in message news:174201c46f52$893379b0$a401280a (AT) phx (DOT) gbl... Thank you for the response: Which Task? Transform data task. Why not use TSQL's BULK INSERT command No idea wether this would be faster than DTS... Does the destination table have 1. Indexes Yes, clustered only. 2. Triggers None. Is the destination DB in at most BULK LOGGED mode? the db is in Simple mode, I'm guessing it's not bulk logged. re you doing this in batches BULK INSERT (BatchSize =) Nope, straight usage of the DTS package (db connection, txt file connection and Transform Data Task in between) What about bcp? I thought this is the same engine used by DTS?! Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.konesans.com - Consultancy from the people who know Thanks again Allan. "A.K." <anonymous (AT) discussions (DOT) microsoft.com> wrote in message news:176a01c46f4b$97257020$a601280a (AT) phx (DOT) gbl... Hi: Any help with this would be appreciated. Thanks in advance. Environment: SQL 2000 SP3 all up to date on Win2K Server SP4 all up to date. Issue: Trying to import a very large text file (12GB) into a db table where the data was originally from. The data was exported based on a selective query (with a WHERE clause), then the table was truncated, and I am in the process of reimporting the data. The idea was to delete a bunch of rows as fast as possible. The export of the text file took about an hour, the import is taking six hours and still running. When I looked at the performance counters, it seemed to be writing and reading to the drive that holds the tempdb. I assumed that its writing the data to the tempdb db for the single shot insert - mind you I did not optimize the package by tweaking the switch in the package properties pages. I am now running out of time, I have to have the data in by ...hmmm.. half an hour ago. Would it be worth the while to wait for it or restart the whole thing with optimized package options for this type of import (which I have no idea at this point how to do, BOL maybe)? Thank you for your help. If you could CC me zebdi (AT) hotmail (DOT) com A.K. . . |
![]() |
| Thread Tools | |
| Display Modes | |
| |