Comments on: 10 Myths of Virtualization – Busted http://www.thevirtualcircle.com/2008/07/10-myths-of-virtualization-busted/ WordPress site Wed, 16 Nov 2011 20:21:06 +0000 hourly 1 By: Admin http://www.thevirtualcircle.com/2008/07/10-myths-of-virtualization-busted/comment-page-1/#comment-355 Admin Fri, 01 Aug 2008 21:02:30 +0000 http://havemacwillblog.com/?p=595#comment-355 Thanks for the contribution. This was a technical nuance I was unaware of. Thanks for the contribution. This was a technical nuance I was unaware of.

]]> By: Rachel Chalmers http://www.thevirtualcircle.com/2008/07/10-myths-of-virtualization-busted/comment-page-1/#comment-354 Rachel Chalmers Fri, 01 Aug 2008 17:40:17 +0000 http://havemacwillblog.com/?p=595#comment-354 "VMware had the neat idea of making this available on the commodity Intel platform..." To be fair, this wasn't just a neat idea. It was a pretty brilliant hack on Mendel's part. People had always assumed x86 couldn't be virtualized because it wasn't <a href="http://en.wikipedia.org/wiki/Popek_and_Goldberg_virtualization_requirements" rel="nofollow">Popek Goldberg compliant.</a> Mendel's insight was that the conditions in which the sensitive-but-not-privileged instructions executed were actually finite, and could be handled in software workarounds. Kudos to him! “VMware had the neat idea of making this available on the commodity Intel platform…”

To be fair, this wasn’t just a neat idea. It was a pretty brilliant hack on Mendel’s part. People had always assumed x86 couldn’t be virtualized because it wasn’t Popek Goldberg compliant. Mendel’s insight was that the conditions in which the sensitive-but-not-privileged instructions executed were actually finite, and could be handled in software workarounds. Kudos to him!

]]> By: Frank@MO http://www.thevirtualcircle.com/2008/07/10-myths-of-virtualization-busted/comment-page-1/#comment-356 Frank@MO Fri, 01 Aug 2008 08:32:46 +0000 http://havemacwillblog.com/?p=595#comment-356 Great post, Robin, very interesting. Your comment that, "computing has been driven by the user interface ever since the first user was invented," struck me as quite profound. With regard to your sixth myth -- reducing the problem of managing servers -- Managed Objects couldn't agree more. Our CTO argued in part, that a service or dependency model is perhaps a prerequisite to server virtualization for this very reason. His article can be found here: http://www.gridtoday.com/grid/2199895.html Great post, Robin, very interesting. Your comment that, “computing has been driven by the user interface ever since the first user was invented,” struck me as quite profound.

With regard to your sixth myth — reducing the problem of managing servers — Managed Objects couldn’t agree more. Our CTO argued in part, that a service or dependency model is perhaps a prerequisite to server virtualization for this very reason. His article can be found here: http://www.gridtoday.com/grid/2199895.html

]]>