In response to my article yesterday about network convergence, Don Marti over at LinuxWorld responds that he is all for convergence.  But he argues, why not converge on a 2 to 4k box, rather than a 10k Cisco box.  Amen to that Don! On the Network Cisco Subnet blog, after rehashing Don's and my positions, the point made is that:

The point of convergence is to save money, as well as to ease administration. At the point where it costs more money or requires more admin than the "old way" of doing things, network pros will have a hard time swallowing it.

I guess they are referring to converging more functionality on one box, you could make administration more complex thereby negating the potential cost savings. I agree.  That is one of the biggest things we have been working on the Cobia platform. How to make managing these diverse applications easier and more efficient.

Back to Don Marti's comments on cheaper boxes though.  There are actually a few rising tides that are floating the convergence boat.  The vastly increased power of off the shelf hardware at those prices is the true enabling technology. Having a cheap box does no good if it doesn't have the horsepower to get the job done.  At the end of the day, that is what kills the 10k Cisco box.  There is no need to pay 10k for the power that the box has when more powerful boxes are cheaper.  The caveat though is, how long do you think it is going to take Cisco to realize that too?

We have contemplated all of these factors in our strategy around Cobia.  We think virtualization is another key driver in this convergence revolution.  Also, by distributing source code with the product, allowing for 3rd party innovation and collaboration, we can leverage a wider community to speed development.

Linux as the common OS underlying much of the convergence trend is a key driver, but there are other forces at play that ensure that we will continue to see consolidation and convergence in the months and years ahead.