[22720] in North American Network Operators' Group
RE: ATM Switch Interoperability Fixed!
daemon@ATHENA.MIT.EDU (Webb, Tremayne)
Tue Jan 19 15:59:57 1999
From: "Webb, Tremayne" <TWebb@mercury.balink.com>
To: "'Alex Bligh'" <amb@gxn.net>
Cc: "'nanog@merit.edu'" <nanog@merit.edu>
Date: Tue, 19 Jan 1999 15:33:04 -0500
It appears that the FORE ASX1000 switches and Ascend Switches do some
manipulation of the settings you give them. The Ascend switches allocate
[SCR (in cells)] * 1.35 as your CAC (Connection Admission Control) and
leaves the PCR untouched. The ASX1000 allocates it as [SCR (in cells)] *
1.75 and leaves the PCR untouched. We recalculated the SCRs to equal out
and this solved the problem.
~~~~~~~~~~~~~~~~~~~~~~~~~
Trey Webb
Sr. NOC Engineer
BA Global Networks, Inc.
703-295-4407
~~~~~~~~~~~~~~~~~~~~~~~~~
-----Original Message-----
From: Alex Bligh [mailto:amb@gxn.net]
Sent: Tuesday, January 19, 1999 12:08 PM
To: Steve Murray
Cc: Martin, Christian; danny@qwest.net; nanog@merit.edu
Subject: Re: ATM Switch Interoperability
Steve,
Formula looks great (interesting). However certainly (Cisco) AIP's don't
traffic
shape sufficiently accurately for (Cisco) BPX's not to cause the
occassional discard. I figured Lightstreams might be the same.
In our case the relevant supplier changed CDVT and it fixed this.
--
Alex Bligh
GX Networks (formerly Xara Networks)
