• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 445
  • Last Modified:

Nightly Imports using DTS or Incremental Updating

I am trying to update a table using DTS.  Here is the setup

1. AS/400 Machine with DB/2
2. MS SQL 2000


I am actually updating invoice history, since querying against the as/400 takes too long, so I decided to move the history files over to the SQL Server to improve performance.  The history dates back quite a bit.

There are errors in the dates, ie. something could have been invoiced July 2006, (the year is wrong) ... As my original idea was just to select the max date on the SQL Server, then select all invoices past that date on the as400.

The invoice numbers are also erronous.

So, I am left with two choices.

1. Complete import of the table during off hours ( 2:00 am)
2. figure out some way of incremental updates. The best solution.

Can someone walk me through how to create/deploy a DTS package that can do this?  Or, if there is a better way than DTS please let me know.
  • 3
  • 3
1 Solution
How much information are you talking about?  You said the invoice date can be wrong--is there an add date or a change date on the records?
2xhelixAuthor Commented:
Hi Arbert!

Once again, thanks for your help.
From what I can tell , the as400 always appends the data to the end of the table.  
There is no add date, change date, .. only invoice date =(

The information is approxmately 3 years worth of data .. approx 700,000 rows.
A complete import takes approx ~6 mins

I tried to

Which to me, is selecting all invoice numbers that are inside the as/400 which are not inside the SQL Server.  This query took ~15 mins?

So I am not sure which is the best solution.  
Well, to get the 6minutes down, you're probably better off to insert EVERY record into a local sql table and then apply your NOT EXISTS logic (don't use not in--query plan isn't quite as good).

If you can't guarantee the accuracy of the invoice number and invoice date, I really don't see a way around it.  Will the 700,000 grow much larger, or do they keep only a certain time period in this table?
A proven path to a career in data science

At Springboard, we know how to get you a job in data science. With Springboard’s Data Science Career Track, you’ll master data science  with a curriculum built by industry experts. You’ll work on real projects, and get 1-on-1 mentorship from a data scientist.

2xhelixAuthor Commented:
700,000 will continue to grow over time.

To be quite honest, I am not really familiar with DTS and what it can provide.  
If I did a nightly import, that would require me to drop the table, and then import the table from as400?

since I believe the invoice history on the as/400 appends to the end of the table, and there are never any updates to any rows, is there some way I can just rowcount (as/400) - rowcount(sqlserver) .. then somehow start a cursor at that location and copy?

You could, but you CAN'T be guaranteed the order of the rows when you get them.  Only by adding an ORDER BY to your select statement are you guaranteed order--might not match the original AS400 table...

I don't think I would drop the entire table--transfer the entire as400 table into a sql server temp table (only transfer columns that you need) then, use the code (like you have above) to figure out what records don't exist and do an insert from the temp sql table....You'll find it's probably faster to transfer every row and put the criteria on SQL Server after the fact--linked servers aren't that efficient with criteria.
2xhelixAuthor Commented:
Great, I just put together everything, and started to import/update several tables last nite, and it is working perfectly.  And the execution times are acceptable as well.

Thank's for all your help .. don't know where I would be in this project without you =)
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Get your problem seen by more experts

Be seen. Boost your question’s priority for more expert views and faster solutions

  • 3
  • 3
Tackle projects and never again get stuck behind a technical roadblock.
Join Now