RE: Data Pump and compress on the fly in Unix
- From: "William Wagman" <wjwagman@xxxxxxxxxxx>
- To: <mwf@xxxxxxxx>, <rjamya@xxxxxxxxx>
- Date: Wed, 10 Jan 2007 15:44:13 -0800
Mark,
Check the Oracle Database Utilities Manual 10g Release 2 (10.2) part
Number B14215-01 Chapter 1, towards the end. The section titled Original
Export and Import versusData Pump Export and Import explicitly states
that Sequential media such as tapes and pipes are not supported. Given
that the data pump can be parallelized it seems to make sense. Then
again, given that the parallel parameter only works in the Enterprise
Edition.
Thanks.
Bill Wagman
Univ. of California at Davis
IET Campus Data Center
wjwagman@xxxxxxxxxxx
(530) 754-6208
________________________________
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Mark W. Farnham
Sent: Wednesday, January 10, 2007 3:20 PM
To: rjamya@xxxxxxxxx
Cc: mark.powell@xxxxxxx; oracle-l@xxxxxxxxxxxxx
Subject: RE: Data Pump and compress on the fly in Unix
Hmm, I've never noticed that note - is it the kind of not supported as
in "don't call us for help in using named pipes" or the kind of not
supported as in "if you use this it won't work"
I'm not even convinced Oracle's software can tell the difference between
a stream of bytes from a named pipe and a stream of bytes from a file.
That transparency is why Kemeny and Kurtz called them communications
files and noted that every data source and every data sink is a file.
I'm moderately sure D. Ritchie followed that lead when building UNIX,
but sometimes I mis-remember things.
Another possible confusion is whether Oracle's note was about Unix named
pipes or what a company convicted of restraint of trade also called
named pipes.
Have you got a reference to that note? I'd sure like to read it and try
to figure out when in the gap between 1988 when Oracle specifically
recommended using named pipes to me and now they desupported them. Of
course that was export, not datapump, but I still can't quite grok how
their software would be able to figure out the difference in the source
of the bytes.
Regards,
mwf
-----Original Message-----
From: oracle-l-bounce@xxxxxxxxxxxxx
[mailto:oracle-l-bounce@xxxxxxxxxxxxx]On Behalf Of rjamya
Sent: Wednesday, January 10, 2007 5:33 PM
To: Mark W. Farnham
Cc: mark.powell@xxxxxxx; oracle-l@xxxxxxxxxxxxx
Subject: Re: Data Pump and compress on the fly in Unix
yes and oracle has a note saying that 'it is not supported'.
Rjamya
On 1/10/07, Mark W. Farnham <mwf@xxxxxxxx> wrote:
Have you considered or tried named pipes?
Other related posts:
