When organizations put a lot of eggs in one basket – desktop side

http://www.zdnet.com/blog/virtualization/when-organizations-put-a-lot-of-eggs-in-one-basket-desktop-side-of-the-story/2103?tag=nl.e539

This is a chicken-little story.

We’ve been putting many computer eggs in one hardware basket for a long, long time.
What do you think mainframes running MVS and VM/CMS were?
What were things like air traffic control?

The ‘desktop’ is a fuzz concept that gets confused with a GUI.
Those mainframes – think airline ticket and reservation – could handle many hundreds of remote terminals, keeping them updated.

What’s a dumb terminal if not the ultimate in ‘thin clients’?

And many tasks? Heck, that’s what UNIX on a PDP-11 back at the beginning of he 80s was doing, and doing damn well.

And restart in case of a crash? Look to IBM again, or look to applications such as the VI editor under UNIX and the – sorry, I forget its name – editor under VMS. Again the early 80s. I know those both
first hand and they had session recovery. As or the mainframe, my experience is limited there, but again I’ve used editors and development environments that retained state across shut-down.

If it comes to it, my laptop has a ‘hibernate’ function. The RAM gets written out to a file or the ‘swap’ segment. Restart or wake-up restores it. The issue is how fast can the write occur – another thing
that SSD impacts! But it doesn’t take much to alter the kernel so that the virtual memory system is a bit like the old ‘roll-in/roll-out’ version of swap and a complete image is always in swap.

Of course if you have a server with 128G of memory those are probably not great paths to pursue, but this article is talking about desktops.

But on the whole this is a poor article because the author so clearly ignores history.