Hi,
I'd like to know if it's possible to create a new protocol akin to tcp, but without its inherent disadvantages. Would it be better to evolve a radically new one, or modifying a present, but more efficient standard, like UDP to achieve the goal? Please advise.

You may have an uphill struggle to replace TCP - it's pretty well entrenched!

I know. I'm not thinking of replacing it. Just creating an alternative.

what you think , which portion of tcp can be improved ??

I'd say that it is definitely possible. However, I will also observe that you did not pose any constraints such as time, cost, compatibility with existing protocols, etc. Consider all the technology that is layered on top of TCP at the moment. Layers and layers of other protocols like HTTP, DHCP, FTP, SSH, etc. Maybe some of the protocols themselves don't rely on TCP specifically to be their transport, but you can be pretty sure that implementations for those protocols would need to come from somewhere!

There is an old hacker saying: "God created the world in seven days, but He didn't have an installed base!"

I'm not seeking any compatibility with the present protocols to achieve it. And is it not possible to abstract the upper layers so that they don't get affected by the change?

If memory serves me right, this has already pretty much been done. I guess TCP doesn't scale well over really high speed connections, so some standards body picked a replacement protocol from about five competing research protocols. I'd post a link or something, but I'm not exactly sure what to google for.