<div>Brian,</div>
<div>I think the ratio of performance cost (for the VM) to performance capacity (of the core hardware) is tending to zero over the long haul.</div>
<div>Recall the time when people said "distributed computing would be trivial if you could boot unix on every node, ha ha"? It was a reductio ad absurdum. Then.</div>
<div>OTOH what Moore Giveth, Gates Taketh Away. But maybe not entirely with the MiniWin.</div>
<div>Peter<br><br> </div>
<div><span class="gmail_quote">On 10/22/07, <b class="gmail_sendername">Brian D. Ropers-Huilman</b> <<a href="mailto:brian.ropers.huilman@gmail.com">brian.ropers.huilman@gmail.com</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">On 10/22/07, Robert G. Brown <<a href="mailto:rgb@phy.duke.edu">rgb@phy.duke.edu</a>> wrote:<br>> This in turn moves their system closer to being a suitable host for VMs,
<br>> as I think the ideal towards which OS's are moving is the kernel as a VM<br>> host, everything else as a VM guest on top of a device-independent<br>> layer. At least I hope we are going there.<br><br>Interesting comment for this particular list. While I'm all in favor
<br>of MS "seeing the light," so to speak, your comment on "... everything<br>else as a VM guest on top of a device-independent layer. At least I<br>hope we are going there." With everything so performance oriented in
<br>our world, I just find this comment surprising.<br><br>--<br>Brian D. Ropers-Huilman, Director<br>Systems Administration and Technical Operations<br>Minnesota Supercomputing Institute <<a href="mailto:bropers@msi.umn.edu">
bropers@msi.umn.edu</a>><br>599 Walter Library +1 612-626-5948 (V)<br>117 Pleasant Street S.E. +1 612-624-8861 (F)<br>University of Minnesota Twin Cities Campus
<br>Minneapolis, MN 55455-0255 <a href="http://www.msi.umn.edu/">http://www.msi.umn.edu/</a><br></blockquote></div><br>