[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GT-ITM Question (fwd)



I'm not sure why this mail failed to get posted yesterday.
I'm trying to send it again,

--Padma.


---------- Forwarded message ----------
Date: Tue, 7 Jul 1998 13:56:26 -0700 (PDT)
From: Padmaparna Haldar <[email protected]>
To: Christos Papadopoulos <[email protected]>
Cc: [email protected]
Subject: Re: GT-ITM Question (fwd)


On Thu, 25 Jun 1998, Christos Papadopoulos wrote:

> Anyone knows how to parse the output of GT-ITM?
> Specifically, I want to distinguish transit and stub nodes.
> I suppose the T and S preceeding a vertice is what distinguishes
> them, but it's not clear what the rest of the items are.


Yes, T indicates a Transit. "T:0.0",A136,53,0 would mean a transit node 0
in the zeroth domain. I'm not sure what the rest means coz I didn't need
them.
and "S:0.0/1.0",A19,71,25 indicates a stub node connected from a T:0.0
transit node and its own nomenclature being 1.0 - zeroth node in cluster 1
in that given stub.

> 
> Also, does sgb2ns preserve the order in the GT-ITM output
> when creating nodes? In other words, if there are 4 transit
> nodes, are these guaranteed to be n(0) - n(3)? It looks like
> the answer is no.
> 

sgb ts graphs are typically created with transit nodes first followed with
stubs, for a given domain, and replicates this for every domain in a
multidomain topology.
sgb2ns doesnot preserve this T-S concept. however sgb2hierns, the
converter used to create hierarchical topologies does preserve the concept
of T-S and domains. 
You can download both sgb2ns and sgb2hierns pkgs from
http://www-mash.cs.berkeley.edu/ns/ns-topogen.html

Hope this helps,

Thanks,
--Padma.

> Thanks,
> 
> Christos.
> 

------------------------------------------------
Be true to your work, your word, and your friend.
--Thoreau

Padmaparna Haldar
ISI/USC.
310.822.1511 #352