bandwidth question

[Tony Hammitt]

In practice, this is being done for most of the network traffic.
As you may know, asynchronus transfer mode (ATM) uses 53 byte
'cells' with (I think) an 11 byte header, so they always transmit
64 bytes.  Thing is, they can transmit an almost arbitrarily
large number of cells, the only limits being the medium and switch
used.  If you want more bandwidth, you just buy better parts but
the cells themselves don't change.  OTOH, ethernet packets look 
different depending on how they are transmitted.

This makes ATM a lot better for cross-country connections and for
thinks like telephone conversations.  Queueing a cell every 10ms
is really simple on ATM, whereas on ethernet, with its bigger
packets, you can't guarantee that your tiny little packets will
get sent at any particular time.  TCP doesn't even guarantee that
it will send them within a reasonable amount of time, prefering
to get more data to send all at once (unless you tell it otherwise).

Back to the point, most WAN ATM is over fiber and most LANs use
copper, just because they are cheaper that way.  Fiber is too
expensive to run around to every desktop and takes more complex
technology to interpret back and forth into electric signals.
Copper is cheaper to use but I think that fiber is probably
cheaper to make.  It's easier to 'mine' sand.

In short, all of this complication is typically hidden from the
users and is usually hidden from the network engineers, too.

BTW, you can send a hell of a lot more bandwidth down fiber than
copper.  Each frequency of light being good for many gigabits
and you can put lots of frequencies on the same fiber.

Hope this helps,

Tony Hammitt

Evan Hoff wrote:
> 
> i had a question come up in my mind recently and was
> curious about the answer..my question is this...
> 
> if you took a piece of fiber optic and a piece of copper (same length)
> and figured the speed of trasmission based on the resistance
> of the gauge of wire and speed of light, then put it into a ratio...
> for an example, we'll say that the speed of light
> (for the fiber optic) was twice as fast as the
> electricity flowing over the piece of copper, in this example
> we would have a ratio of 1:2
> so..if we setup a piece of fiber, and broke down
> all the packets (A and B) transmitted into two smaller units
> (A1, A2, B1, and B2) and then transmitted them alternating
> between packets A and packet B. (sending A1, then B1, A2, B2)
> which would essentially increase latency for each user, but
> at the same time, allow two users to transmit at the same time.
> appearing to run at the copper speed for both (2:1)
> essentially halfing the speed of fiber and doubling the bandwidth...
> in theory...would the overall network load be better (if the
> network segment were supporting a large user base) or would
> the result be a complicated network that would run at the same
> speed as the original?  note: im only speaking in theory, not
> practical..
> 
> Evan Hoff
> evanh23 at usa.net
> 
> ____________________________________________________________________
> Get free email and a permanent address at http://www.netaddress.com/?N=1
>