Ascend GRF

Nathan Stratton nathan at
Wed Jun 11 15:49:12 UTC 1997

On Wed, 11 Jun 1997, Peter Leppik wrote:

> I've been hearing conflicting stories about the Ascend GRF lately.
> Cisco claims that it doesn't really work as advertised in "realistic
> networks," but Ascend says this is because Cisco's routers melt down
> when the GRF feeds them routing table updates, so the GRF has to be
> throttled back to work in a mixed environment.
> Does anybody have any real experience here?  Any information, pointers,
> anecdotes, or wild rumors would help.

Yes, right now we have 100% GRFs in our network, and have had a few
problem with more then 46 peering sessions on one GRF. This problem was
fixed Friday, and we now have over 50 sessions up on our MAE-East router.
We also ran into problem with people sending us a router ID of
(not something they should do), Ascend added a flag to gated to let gated
peer with if you want. 

So far we are happy with the GRF, there are some thing I would like to
see. I think there is a need for redundant processor cards, and the
ability to load balance over interfaces. I still think the GRF blows away
the Cisco 7500, but we have not tested the new RSP4 cards. We plan to test
this in the next few weeks, and will then decided if we are going to say
cisco free, or if we need to make a hardware change.

Nathan Stratton                                President, NetRail,Inc.
Phone   (888)NetRail                           NetRail, Inc.
Fax     (404)522-1939                          230 Peachtree Suite 500
WWW                Atlanta, GA 30303
"No king is saved by the size of his army; no warrior escapes by his
great strength.                                        - Psalm 33:16

More information about the NANOG mailing list