[ale] parallel processing

Joseph A Knapka jknapka at earthlink.net
Tue Jan 8 05:18:41 EST 2002


"Glenn C. Lasher Jr." wrote:
> 
> Try something like this:
> 
> for FILE in pattern_to_be_matched
> do
> cp ${FILE} place_to_copy_to &
> done
> 
> I do this with gzip jobs at work (usualy 30+ files at a rip) in order to
> take full advantage of the 4-processer Alpha we are running.

It seems this will gain you little, since as has been mentioned,
disk copy is I/O bound. Those four processors are all trying to
shove their data down the same pipe to the disk, a pipe which
is easily saturated by a single processor.

Cheers,

-- Joe

> On Fri, 4 Jan 2002, jeff hubbs wrote:
> 
> > Andy -
> >
> > You can execute each cp command with an ampersand ("&") at the end
> > (that's sometimes called "amping off") but unless the files/dirs are
> > really big, the cps will finish before you can type the next one - is
> > that what you're talking about?
> >
> > - Jeff
> >
> > Zyman, Andy wrote:
> >
> > > Hello,
> > > I just wonder :
> > > Let's say in Oracle we have parallel SQL processing - I can specify that
> > > certain SQL should be processed in parallel.
> > > What about Unix? Let's say I have  a directory on Sun server with 10 files.
> > > I want to cp them to the diff. one. I'm doing cp xxx yyy. Now how it will
> > > processed and who I can see that it will be done in parallel?
> > >
> > > Thank You
> > >  Andy
> > >
> > > ---
> > > This message has been sent through the ALE general discussion list.
> > > See http://www.ale.org/mailing-lists.shtml for more info. Problems should be
> > > sent to listmaster at ale dot org.
> > >
> > >
> > >
> >
> >
> >
> >
> > ---
> > This message has been sent through the ALE general discussion list.
> > See http://www.ale.org/mailing-lists.shtml for more info. Problems should be
> > sent to listmaster at ale dot org.
> >
> 
> --
> glasher at nycap.rr.com
> Ask me about how Sprint is willing to lose customers over 7 cents.
> 
> ---
> This message has been sent through the ALE general discussion list.
> See http://www.ale.org/mailing-lists.shtml for more info. Problems should be
> sent to listmaster at ale dot org.

-- 
"I should like to close this book by sticking out any part of my neck
 which is not yet exposed, and making a few predictions about how the
 problem of quantum gravity will in the end be solved."
 --- Physicist Lee Smolin, "Three Roads to Quantum Gravity"

---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.






More information about the Ale mailing list