Packet Generator

i designed a relatively complex atm network, and it took me about four days to complex building it on OPNET. and after runnign the simulation for a week (read 4 minutes in real time on OPNET), while trying to retrieve some critical statistics, i realized i had not setup any of them. how brilliant of me!!!!!!!

now i have to set up all the statistics i need to observe and then i will have to run it again. but this time i will run the simualtion for a month. should take about 5 to 6 minutes on my pc.

found so many flaws in IP v4, so many options not used, reserved or maybe even forgotten, i do agree IP v6 is protocol made after keeping in mind all the lapses in v4, but even then, i found some areas, which could be avoided. i would put them on as my next post

We'll all be looking forward to that Arani

IMO, IP wasn't so much the problem as TCP. IPv4 is a fairly decent protocol -- the unfortunate part was its horrendous implementation in some network stacks.

I expect IPv6 to be too much -- IPSEC is all very well, but when you complicate something, you're destined to have a problem. There are now so many more things for developers to keep track of.

Cheers,

sorry for the delay gentlemen,
i studied the various protocols, specially IPv4 and v6, found out their loop holes, places where there were options unused or plainly reserved where IETF did not know what to do with them. anyway, i have managed to design a new protocol, completely different from what we see in conventional networks, but at a closer look, its got the best of IP and ATM, regarding various factors like PMTU, congestion control, fragmentation control, loose circuit formation for data transfer, much more converstion friendly interface for AAL5 if it would be used across conventional ATM networks, and much more. I have created a link model, a node model, a process model, and a load structure to test it. the virtual network has already been created, and the protocol would be implanted onto the corresponding network entities before I go ahead for a full scale simulation. at first hand it might fail, but it will still give me a direction.
more to come.........thanks for waiting

wow, sounds like you're making some good progress. This may or may not help you, but for your packet generating and recording/analysis needs, you might want to look to Ixia at www.ixiacom.com ... They're hardware packet generators are very cool, although also very expensive. But, if you've got some funding from a University you never know. Also, if you find the right reseller you might be able to get a loaner for 'evaluation'.