In SQL Server 2008, I am trying to BULK INSERT CSV about 1 GB in size. Since I am doing this, it creates a huge tempdb.mdf file. Right now, this is 35 GB for this 1GB CSV file.
I have tried various solutions that Microsoft provides and they don't seem to work.
I think the easiest way to “compress” the tempdb.mdf file on a non-production system is to simply delete it while the SQL service is down.
Will this cause problems? If so, what problems can be expected?
edits
1) Here is a line from CSV (has about 4M lines):
PS D:\> gc .\map.items.csv | select -last 1
40747646;jdbc:patent/8046822;8683;other/patent;12/31/69 16:00:00.00 PST;E6 E6 80 6D FD 6D 0B 5F 44 66 4E 1C 35 DE 46 BB 19 36 3C 31 37 67 4D 1D DF 58 A5 99 A8 A0 3B 52;crawled;full_patent_db2;Electronic apparatus, function selection method of electronic apparatus and management system of electronic apparatus;Sony Corporation;Tokyo;03;G06F21/00
2) Here the database describes the information about the table (nothing exotic and triggers): https://gist.github.com/mlissner/4cd13db5a1bbae91dd50
3) I have a database configured for a simple recovery model.
source
share