[SI-LIST] Re: Fibre channel interconnect margins

  • From: "Andrew Ingraham" <a.ingraham@xxxxxxxx>
  • To: <si-list@xxxxxxxxxxxxx>
  • Date: Tue, 4 Jul 2006 09:36:29 -0400

> Based on raw testing with PCI Express, you would need 67 minutes to test
> a BER of 10e-13, 11 hours for a BER of 10e-14, and 4.6 days for a BER of
> 10e-15.

I've not done this sort of thing, but I think it's a lot more than that.
Don't forget, these are random processes (or at least, the mathematics we
use to describe them are).

It isn't sufficient to run the experiment just long enough for (on average)
one error to occur, and then if the number of observed errors is either 0 or
1, claim victory and go home.  The BER could be greater than your limit and
yet the particular interval you chose had only 0 or 1.  Or you might observe
2 or 3 errors in that particular interval even though the BER was what you
thought it should be.

Personally, I'd want to run the experiment over at least 10 times as long an
interval, to have enough confidence that my measurement is even close to the
statistical average.

Which is why you are totally correct when you said "a low BER means a LOT of
testing."  An even bigger LOT than you thought it was.  Some experiments
just aren't practical, as written.  They need to be re-written (if you are
clever enough to know how) to make the errors happen faster.

Personally, I am of the school of thought that most real processes of this
nature may look random and/or gaussian, but that it is only a mathematical
approximation that works pretty well AS LONG AS we don't go to extremes and
look in every dark corner.  Sorry, Steve, I disagree with you that real
physical noise is truly random and unbounded.  It just behaves that way,
99.9999% of the time.

F=ma works too, as long as you don't hold the universe to it....


To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List FAQ wiki page is located at:

List technical documents are available at:

List archives are viewable at:     
or at our remote archives:
Old (prior to June 6, 2001) list archives are viewable at:

Other related posts: