Hello all @ 3000-l from Flagstaff Arizona!

Re: [HP3000-L] :DSCOPY versus FTP performance

I found a wireless hot-spot in the lobby of the Quality inn in Flagstaff and thought I could do a quick reply between adventures at the O.K. Corral, Grand Canon and a Jeep ride in Sedona.

Over the holidays I did a bit of performance testing between 2 MPE systems to compare DSCOPY vs FTP.  I will publish my results to the news group early in January when I return to Atlanta and get a chance to verify my results.  I tested with 80 byte and 258 byte binary, ASCII and byte stream of size 1gb, 2gb and 5gb.

My test were between a 989-400 and a 987-200 both with maximum memory over an isolated 100bt full duplex link.

Your Mileage Will Vary (YMWV),  but in general I found that FTP was 30-40% faster than DSCOPY in  binary 3000 to 3000.  FTP was just as fast with  byte stream files as binary.  DSCOPY of course could not transfer byte stream.  FTP was 30-40% slower than DSCOPY in ASCII 3000 to 3000.  The ASCII slow performance is expected since the 3000 FTP/iX reads in record mode to support inserting the record deliminator to non-3000 systems.

Note:  I do expect DSCOPY will win in cases where file record lengths decrease to less than 80 bytes....   The smaller the record length, the poorer FTP/iX will perform against DSCOPY.  DSCOPY transfers blocks.  FTP transfers records in mapped file mode for most file types.

Well, I am on to Phoenix tomorrow... We will see if I can find the Internet at my next destination.  I hope this helps and I wish you all a happy holiday season!






Regards,

James Hofmeister
Email: <first>.<last>@hp.com
Hewlett Packard - Global Solutions Engineering (WTEC)
P.S. My Ideals are my own, not necessarily my employers.

---------------------------------
Do you Yahoo!?
Free Pop-Up Blocker - Get it now

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *