And we thought the text part of the Starr Report would be bad
David Stoddard
dgs at us.net
Fri Sep 18 21:39:35 UTC 1998
Paul Zawada writes:
>[snip ...]
> You can only fit 1.536 Mbps of real data down a T1 unless you've figured
> out a funky way to send it across using the framing bits.
>
> 8 b per channel * 24 channels per frame * 8000 frames per s = 1536000 bps
> 28.8 kbps * 53 = 1.526 Mbps
> 28.8 kbps * 54 = 1.555 Mbps
>
> So technically they should have used the numbers 52 and 53. :-)
There is more to this than meets the eye -- 28.8K is asynchronous
and has start and stop bits for every byte, so there are a maximum
of 2880 bytes/sec available over 28.8K. Then there is the issue
of latency and IP overhead which tends to approximate 89% of the
available bandwidth, or 2880 x .89 = 2563 bytes/sec. The T1 circuit
normalized to bytes/sec is (1536000 / 8) = 192000 bytes/sec, so
accounting for apples and oranges shows 192000 / 2563 = 74.91 sessions.
Ofcourse, this assumes that all 75 28.8K modems are pulling data
in the same direction at the same time. In reality, traffic for
most ISPs runs about 4:1 to 5:1 inflows vs outflows. Because T1
circuits are full duplex (1536000 bps in two directions at the
same time), and assuming a 4:1 ratio of inflows to outflows, this
would allow 75 + (75 / 4) = 93 28.8K connections. Fortunately,
customers don't actually saturate the lines the entire time they
are connected -- some actually pause to read the screen once in
a while -- so the real usage on the line has a tendency to
approach 8:1, 10:1, or even more sessions. With a conservative
number of 8:1 inactive to active sessions, 8 x 93 = 744 sessions.
The Starr Report and other nonsence have a tendency to blow
these types of calculations out of the water every now and then,
but then no provider has built their network on the worst case
scenareo. YMMV.
Dave Stoddard
dgs at us.net
More information about the NANOG
mailing list