Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!cs.utexas.edu!bcm!lib!mdaali.cancer.utexas.edu!drg
From: drg@mdaali.cancer.utexas.edu (David Gutierrez)
Newsgroups: comp.protocols.appletalk
Subject: Re: Request for optimum EtherTalk Range setting
Message-ID: <4586@lib.tmc.edu>
Date: 22 Jan 91 17:03:04 GMT
References: <36032@netnews.upenn.edu>
Sender: usenet@lib.tmc.edu
Organization: Univ. Texas M.D. Anderson Cancer Center
Lines: 31
Nntp-Posting-Host: 129.106.3.196

In article <36032@netnews.upenn.edu> tony@scotty.dccs.upenn.edu (Anthony 
Olejnik) writes:
> I don't expect to have more than 1000 Ethernet (EtherTalk) connect
> Macs on campus.  Taking this into account, what should we set our
> ethertalk range to (in our AppleTalk routers)? 
>
> How about 1-4?  This would give us a max ATPH2 net number of 4.
> 4 times the max ATPH2 node number (which I think is 253) is 1012.

This should work. I don't know if there will be problems as you actually 
approach the 1012 node limit. We set ours from 1-10 to allow LOTS of room.

> How about 10-13?  This, too, would give us 1012 total addresses.
> Is this better/worse/doesn't matter than 1-4?

This is the same as 1-4. Makes no difference.

> How about 1-128?  This would give us plenty of addresses for the
> EtherTalk devices to choose from.  Would we suffer any better/worse
> performance this way?

Ack! The documentation for my FastPath warns that the range should not be 
*too* big, because each network number uses another quantum of memory in 
the router. I don't know what kind of routers you're using and if they 
would have memory problems with this, but 128 networks is clearly 
excessive for 1000 nodes.

David Gutierrez
drg@mdaali.cancer.utexas.edu

"Only fools are positive." - Moe Howard