You are wrong. I and many others have been looking for this kind of support
for sometime. I asked a similar question about 6 months ago. If there was
no demand for this kind of thing why did someone go out and hack the
load balancing protocol for Linux.
There are still parts of the world (USA and Canada included) that ISDN is
either not available or simply too expensive.
I am in Manitoba, Canada; a canadian province of a little more than
1 million citizens, of which approx 62% are within the capitol of Winnipeg.
ISDN is only available in Winnipeg. Anyone outside of the captial has to
rely on standard analog telephone lines. Even in the city of Winnipeg, it is
still cheaper to use two analog lines that a single B channel ISDN line.
I do consulting for a number of ISP's who happen to be using Livingston
equipment - Their numbers are 0 in Winnipeg, 9 outside of Winnipeg, thus
my customers who have customers who WANT more than 28.8 bandwidth have to
no choice but to purchase more Livingston equipement as opposed to using
equipment they arlready have - but I guess this is what Livingston really
wants.
I see too options for the Win95/NT users wanting to use 2 analog "bonded"
lines.
1. Get a a replacement driver for Win95/NT that supports Livinston's
multichannel load balanacing. But this will never happen as this is
a "proprietory" protocol. This means Livingston will have to either
develop this driver or release the protocol description. Free drivers
don't generate a revenue stream for Livingston.
2. Build in MLPPP for async channels on ComOS. Now I have heard the
argument that MLPPP requires lots of memory - Is this a valid argument
when the PM2 can be upgraded to 4MB from 1MB for less that $20.
Ask yourself, why has Microsoft integrated this functionallity into there
operating systems. MLPPP is a "standard" now. To stand behind the same aguments
Livingston has made regarding modem standards when with Livingston adopt
a multi line load balancing technique that is standard and allows a
"open systems" concept of plu and play connectivity. Stop the hyprocity
and develop a consistent stand.
I realize that Livingston's engineering resources are probably being taxed
to their limits right now. I recognize Livingston equipment as being
a low $/port effective solution for dialup PPP, but all requests to
add functionallity that makes debugging, management and administration of the
equipment easier gets set aside for higher pressure market demands. I see
a need to develop this functionallity in parallel with thoe demands for
things like OSPF.
Quite frankly, I had written Livingston off on this stuff a while ago when
I realized that to use load balancing or MLPPP in any kind of POP with more
that 30 ports, required mult-chassis support. I beleive today this support
only exists in the PM3. When ComOS for Pm2's will support MCPPP there
may be hope again - but then again by that time it may be too late for
Livingston to haved captured that market oppurtunity.
-- Darren R Besler PO Box 2207 Stn Main Beehive Enterprises Winnipeg, Manitoba, Canada, R3C 3R5 Ph: 204-896-6103 Fax: 204-896-4168 Email: dbesler@beehive.mb.ca