Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: (ASCEND) ATM MTU 64 KB ?
At 04:29 PM 12/23/97 +0100, Tassilo Erlewein wrote:
>No I'm not. I had a discussion with a colleage about wether or not
>64 KB ATM IP MTU is a common feature. In his opinion some Netstar
>people made him believe the GigaRouters could do such a thing -
>two years ago or so. It even seems to be possible if you consider SDH.
>And RF1626 only recommends the 9180 bytes default ATM MTU.
>But I never heard anything about ATM boards/routers supporting
>ATM MTU's higher than 9180 bytes.
>In fact we tried to set the GigaRouter's ATM OC3 to 64 KB MTU.
>(this was some playing around with the SNMP vars...)
>We then saw up to 16 KB IP packets travelling unfragmented through
>the GigaRouter (HiPPI to ATM), just before the GR's crashed ;-)
>Now there is this hint in the manual saying ATM MTU over
>9180 bytes is not supported.
>
>But this Cray ATM OC3 board can obviously do this thing.
>And it would be a nice option in a partially HiPPI-based environment
>in order to avoid HiPPI to ATM fragmentation.
>But again, I'm not sure I want to rely on a 64KB-option in
>the AAL5 standard if nobody (except Cray of course...) ever
>goes higher than 9180 bytes.
As you say, the industry standard seems to be 9180. In fact, that is what
most of the popular chipsets support. I'm not surprised Cray built a
specialized device. Specialized is what Cray does best. :)
Matt Holdrege http://www.ascend.com matt@ascend.com
++ Ascend Users Mailing List ++
To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd: <http://www.nealis.net/ascend/faq>
References: