13 May
1999
13 May
'99
6:23 a.m.
>This very concept came up for lengthy discussion
when I was in college at
>Ohio State. Then, we envisioned using UUCP to store and forward two weeks
>of data,
No hard disks necessarily involved - just the latency
of the network.
You're right, though, imagine using hard disks as cache to increase
the latency, only passing along the packets when the outgoing pipe
seemed empty. Long latency reminds me of early cable modem data
systems, where information was regularly recycled every few hours,
and you could tell your receiving software what to grab the next
time around.
The idea is nice, using the net like a bubble memory, but capacy
is rather small, since any delayed loop storage is determinated
by bandwidth and trip time. Let's just assume we have a 1 Mbps
connections (full dupelx), so we can send (and recive) 128 KBytes/s.
If we now add a delay time for roundtrip of eight seconds (prety
high - 1s would be more realistic), we just get along with 1 Mega-
byte of storage. As you see, the limit itn't the net (and the
possible net resources to use), but rather the bandwidth of our
connection.
And dont tell me you would open more 'storeage connections' to
expand - we have already assumed the full bandwidth of our 'memory
interface' - you can't put more data in.
Ethans solution of derefered methods is more usable, since it
would enhance the latency time - even if we assume a pure memory
based solution (no mail buffer on HD), we can come up with several
minutes of delay (but in fact, all mail servers will use temp HD
storage - but hey, /temp isn't considered as disk - it is just
always to small :).
Gruss
H.
--
Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
Vote against SPAM: http://www.politik-digital.de/spam/en/
Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
Ich denke, also bin ich, also gut
HRK