Re:Tuning JVM for a VM - Lessons Learned, Directly from VMware
by derleth on Thursday April 12, 2012 @08:25PM
(#3834813)
So, how reasonable is it right now to run a JVM directly on a VM?
[ Parent
| on HN ]
Re:Tuning JVM for a VM - Lessons Learned, Directly from VMware
by gaius on Friday April 13, 2012 @04:47AM
(#3835926)
Oracle are running JRockit directly on the hypervisor, no OS.
http://docs.oracle.com/cd/E23009_01/doc.200/e22518/introduct...
Re:Tuning JVM for a VM - Lessons Learned, Directly from VMware
by fleitz on Thursday April 12, 2012 @08:57PM
(#3834887)
Pretty reasonable as proof of concept, MS runs the .NET VM as an operating system, see Singularity. There's also JavaOS, JNode, JX
In practicality the issue is drivers, databases, webservers, etc.
Re:Tuning JVM for a VM - Lessons Learned, Directly from VMware
by derleth on Thursday April 12, 2012 @11:58PM
(#3835334)
There are a few ways to solve this; the simplest might be to replace the OS with a library (a libos) linked into the JVM that implements whatever the JVM needs from an OS API in terms of what a specific VM offers. This is the idea of the MIT Exokernel Operating System, XOK.
Re:Tuning JVM for a VM - Lessons Learned, Directly from VMware
by zokier on Friday April 13, 2012 @04:42AM
(#3835911)
http://labs.oracle.com/projects/guestvm/
edit: see also LiquidVM, possibly aka JRockit VE