Cisco Blogs

Learning from the First Private Cloud

May 21, 2010 - 2 Comments


So, this should be a good Friday post–its short and there are cash prizes involved. 

Well, actually, its the first part of a post.  I have a question for you readers.  I have my own thoughts on the answer, but I wanted to see what other folks thought first, before I taint the discussion.  I have been spending a fair number of cycles on a working group pulling together our company perspectives on cloud computing and where we fit in.  While I was reading something on private clouds, something seemed oddly familiar, which leads me to the question:

Were mainframes the first private clouds?

Bear with me–there are a lot of similarities: based on virtualization, pooled resources, reallocation of resources via policy.  Also, as someone pointed out to me the other day, some of the potential downsides like vendor lock-in, lack of portability, etc. 

So what do you think?  I have a Amazon or iTunes gift card for the best (or most entertaining) argument for and against.  While you are making your argument, tell me what you think we can we learn from mainframe days that can help us today as we look a cloud models?  For example, the mainframe folks certainly had the billing and accounting thing nailed.

So, what do you think?


PS For my next post, I am going to explore if the DEC VT100 was the first instance of VDI? 🙂

In an effort to keep conversations fresh, Cisco Blogs closes comments after 60 days. Please visit the Cisco Blogs hub page for the latest content.


  1. You are correct, most people forget that VM was a mainframe OS that provided for virutalization of IBM mainframe operating systems YEARS before it could be done in intel platforms.

  2. The most terrifying future the cloud presents is not vendor lock in, but enduring jokes like What a dumb terminal!”” again.I think we’ll see local cloud servers, mitigating the lock-in aspect and giving companies more control over their own data. It’s ultimately just a piece of software, it doesn’t take much hardware to handle a few million requests per day, and the barrier to entry regarding coding and testing a bit of functionality is a great deal lower.”