The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Nov> msg00092



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

LSP failure detection

  • From: "Brijesh Kumar" <bkumar@ennovatenetworks.com>
  • Date: Fri, 10 Nov 2000 13:30:18 -0500
  • Importance: Normal



> -----Original Message-----
> To: neil.2.harrison@bt.com; glen.turner@aarnet.edu.au; MPLS Group
> Subject: Re: LSP failure detection
>
>
> Neil, Glen,
>
> As we approach a 40 Gbps network (OC-768), a 50 ms outage
> causes the loss of
> 2 Gb of data. If you consider a bunch of telephone calls
> aggregated on a
> (for example) SONET/SDH path, it's insignificant because the
> loss that any
> one takes is small, even if they have to re-dial. But the
> next generation of
> "television" (beyond HDTV) may well be a 3D imaging system,
> and the amount
> of data in a 3D "frame" is immense, relatively. A consumer
> watching a sitcom
> in 3D can afford to lose a few frames, but a telemedine
> application cannot.
>


Well, you have rightly articulated the relationship between high
transmission speed and the need of lower latency. But remember, that
in a few years time - you would also likely have optical buffers where
data can be temporarily buffered for short duration so that it can be
re-transmitted once alternate path is available.

Cheers,

--brijesh
Ennovate Networks Inc.