Re: Split a dmp file before import
Date: Tue, 10 Nov 2009 07:39:06 -0800 (PST)
Message-ID: <e757a495-d226-470d-af0f-5e86ce1f868a_at_d21g2000yqn.googlegroups.com>
On Nov 10, 10:02 am, gazzag <gar..._at_jamms.org> wrote:
> On 10 Nov, 12:22, Sashi <small..._at_gmail.com> wrote:
>
> > Hi all, I have a dmp file that contains about 9 million rows.
> > Is there a utility/technique that will split the file into two (or
> > more) so that I can run the import in two (or more)
> > stages?
>
> > Thanks,
> > Sashi
>
> 1. What Oracle version?
> 2. What operating system?
>
> Importing from a compressed pipe springs to mind but that's subject to
> your reply to my second question.
>
> HTH
>
> -g
Sashi, I would say the short answer is NO. You can run table= imports to bring only specified tables on a specific import and you can separate bring indexes/constraints from the import of the table data but you cannot split bringing in a specific table in the middle nor run the import without access to the entire dmp file series.
You can use the query= feature to split the export of the data from one table into separate dmp files so that it could in effect be brought back in pieces.
Also you can export partitioned tables by partition and bring back partitions depending on your Oracle version and utility choices.
But once you have a dmp file your options are limited based on how the file was created and what it holds except that with the impdp utility you can also use the query= parameter to filter the input stream. So if you know your data then in theory you could potentially bring partial table contents but you woul d still need the entire complete dmp file (and the dmp file would have had to be produced by expdp)
Generally speaking 9 millions rows is not a lot. More important is the total size in Megabytes, Gigabytes, or Terabytes of data that the 9 million rows represent.
HTH -- Mark D Powell -- Received on Tue Nov 10 2009 - 09:39:06 CST