December 11, 2007 at 1:48 am
Hi Guys,
Does anyone perhaps know of any way you could export data faster from one server to another server.
I used a DTS package to do this. The records in the table in 180 million. I just want to know whether there a a faster way to do this. Someone mentioned to me to use bcp but i have never used this before. Can someone please explain to me how this works.
Regards
IC
December 11, 2007 at 7:14 am
Hi Imke,
Here are a couple links with good articles
http://blogs.msdn.com/sqlcat/archive/2006/05/19/602142.aspx
http://msdn2.microsoft.com/en-us/library/aa196739(SQL.80).aspx
Google this for more: bcp parallel data load
How is the source table architectured. One table with 180 million rows or are the tables behind a partitioned view?
How is your current DTS built? One data pump moving the data?
One thing that you could try is to create several source connections and several destinations with one transformation between each.
The source of each transformation would have a query that selected a range of data by quarter month, whatever makes sense.
Moves Q1 2007
S1------------------------->D1
Moves Q2 2007
S2------------------------->D2
...
Moves Qn yyyyy
Sn------------------------->Dn
Also increase the maximum task in parallel on the Package Properties General tab to a hight number.
Watch the package execution on the desitnation server via sp_who2 or the profiler to make sure there are no blocking issues.
Norman
DTS Package Search
December 11, 2007 at 7:23 am
Hi Nite_eagle
Thanks for the links below....and no the table is not behind a partitioned name.
Regards
IC
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply