[Ns-developers] NS-3 GUI (visualization again)

[Jeremy]

Tom,

I have really been swamped with other things since the suggestion of using
the nam trace. Today, I just quickly glanced over the nam format and wrote
down a list of what we need and tried to reconcile the two. I'll outline my
list of needs and my impressions of how it relates to nam. Let me know if I
missed something or am wrong.

What we need:

    * Initial Node position (wireless)
    * Node movements (including when, where, speed - wireless)
    * Packet Events (time, involved parties, port, packet comment, network
layer)
    * Node energy information (Who is involved)
    * Wired Link existence
    * Wireless link existence (nice, but optional)

Initial node position: nam may have this information, but it is not clear if
it always has this information
Node Movements: nam may have this information, but it is not clear if it
always has this information
Packet Events: nam does have this information. The one thing we don't have
is which network layer the trace is describing.
Node Energy: this seems to be available from the packet events. It is not
clear if it always has this information.
Wired Link existence: nam seems to have this.
Wireless link existence: nam seems to have this as well, though this is not
completely necessary for iNSpect either way.

-Jeremy


On Mon, Sep 1, 2008 at 10:51 PM, Tom Henderson <tomh at tomh.org> wrote:

> Mathieu Lacage wrote:
>
>> On Thu, 2008-07-31 at 13:54 -0600, Jeremy wrote:
>>
>>    1. NS-3 integration is important, obviously. The biggest problem we
>>> have
>>>   here is that there is no fully defined trace format for NS-3 - for
>>> reasons
>>>   I've read in this list. The main thing holding us back at this moment
>>> is
>>>   that we don't have a good understanding of NS-3's inner workings and
>>> don't
>>>   have the personnel to learn it this moment.
>>>
>>
>> If you have specific questions about tracing, posting them here or on
>> ns3-users will likely get you detailed answers. i.e., I (and probably
>> every ns3 developer) would be happy to help you figure out the
>> problematic pieces if it helps make inspect improve quickly.
>>
>> Mathieu
>>
>>
>>
> Jeremy,
>
> I wanted to briefly follow up with you on trace formats for ns-3.  I
> discussed with Mathieu on Friday and we agreed that it would be best if you
> stated your wish list for trace format for iNSpect and we can try to support
> it.  Is there a documented format already?  Or do you want to base a new
> format on extending the existing nam?
>
> - Tom
>
>


-- 
Curiosity didn't kill the cat, it taught it something it didn't know before.