On Feb 18, 2006, at 11:50 PM, Bruce Black wrote:
OK but it is far from *INSTANT* the propagation delay is still there,
no matter how fat (big) of pipline you may have, right?
Ed, I don't quite see your point.
In the technologies that Ron and I described, when you access the
data on the target disk, it either reads it from the target (a
track already copied) or reads it from the source (an unupdated
track not yet copied), the speed is about the same. So when you
access the disk to back it up, the read speed is about the same as
if you were accessing the original disk.
BTW, we are not describing remote copies, to another control unit,
we are describing replication technology from one location to
another in the same control unit. No pipeline involved, just read-
cache-write
Bruce,
I am, most concerned with the "copy that is perhaps miles (up to
1000's). Where I last worked (IIRC) we had approximately 160
volumes that were XRC'd that went from Chicago to Colorado then back
to NY. The (only other) copy being in Poughkeepsie. This is old
information so it might have changed from then.
To me, the information that was bounced around the US was so out of
date it seemed like a waste of effort to do so. I know that XRC and
"dual copy" which (to me) is what you are talking about are similar.
I think the main issue is "distance, IIANM.
Ed
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
