![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
I am having a weird problem when I try to import a text file into SQL Server 2000 using DTS. Only happens I import into an existing table that has various field types, such as nvarchar, float, decimal, datetime, etc., as appropriate for each field. After the import, the table size on disk is VERY large. Where the source file is about 60 mb, for example, the resulting SQL table is 19 gb! This does not happen if I simply import into a table where all fields are nvarchar(8000) and then do the conversion inside SQL Server. When I look at the resulting table, it looks normal -- just takes up way too much space on disk. What am I missing?! Thanks! |
#3
| |||
| |||
|
|
Is the table simply fragmented? DBCC SHOWCONTIG() How are you arriving at the table size? -- Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.Konesans.com |
#4
| |||
| |||
|
|
I am looking at "Table Info" tab in Taskpad. I also checked the size of the physical .MDF file. Importing 172000 records took about 1 hour and increased MDF file size by 19 GB. This is on a new Itanuim-2 box with 64 GB RAM and a super-fast SAN. Thanks for showcontig tip. "Allan Mitchell" wrote: Is the table simply fragmented? DBCC SHOWCONTIG() How are you arriving at the table size? -- Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.Konesans.com |
#5
| |||
| |||
|
|
When importing then people tend to remove indexes. -- Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.Konesans.com |
#6
| |||
| |||
|
|
This table has no indexes. "Allan Mitchell" wrote: When importing then people tend to remove indexes. -- Allan Mitchell MCSE,MCDBA, (Microsoft SQL Server MVP) www.SQLDTS.com - The site for all your DTS needs. www.Konesans.com |
#7
| |||
| |||
|
|
I am having a weird problem when I try to import a text file into SQL Server 2000 using DTS. Only happens I import into an existing table that has various field types, such as nvarchar, float, decimal, datetime, etc., as appropriate for each field. After the import, the table size on disk is VERY large. Where the source file is about 60 mb, for example, the resulting SQL table is 19 gb! This does not happen if I simply import into a table where all fields are nvarchar(8000) and then do the conversion inside SQL Server. When I look at the resulting table, it looks normal -- just takes up way too much space on disk. What am I missing?! Thanks! |
![]() |
| Thread Tools | |
| Display Modes | |
| |