Hi,
I'm interested in understanding the development roadmap for libvirt-cim, what is the
feature set that is provided in the current release and what else can be expected in the
months to come. Moreover, what is a realistic timeframe within which we can see
"libvirt" like functionality being made avaiable through the CIM provider
(actually managing VMs, provisioning, modifying resource allocations and other lifecycle
management functions)?
Would this information be available somewhere?
Regds,
Gaurav
---------------------------------
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
Show replies by date
Hi Gaurav,
GC> I'm interested in understanding the development roadmap for
GC> libvirt-cim, what is the feature set that is provided in the
GC> current release and what else can be expected in the months to
GC> come.
It depends on which platform you're interested in. I think we're
getting pretty close to supporting what libvirt supports for each
platform, at a high level. There are some details that we don't cover
at the moment, like supporting all of the different network types that
KVM supports, etc.
I'm planning to snap another release shortly after Fedora 9 releases
to refresh the version there. It will look mostly like what is in our
development tree right now.
For future releases I have the following high-level goals:
- Scheduling parameter support
- LXC support
- Integration with other physical instrumentation for the HostSystem
and host physical devices
- Proper modeling of the virtual console
- Create/Destroy/Modify of resource pools for network and disk to
correspond to the underlying functions in libvirt
Do you have anything you'd like to see in that list?
GC> Moreover, what is a realistic timeframe within which we can see
GC> "libvirt" like functionality being made avaiable through the CIM
GC> provider (actually managing VMs, provisioning, modifying resource
GC> allocations and other lifecycle management functions)?
Doesn't most of that work already? Defining guests, modifying their
resources, and pushing them through the lifecycle is something we have
been able to do for quite some time now.
Thanks!
--
Dan Smith
IBM Linux Technology Center
Open Hypervisor Team
email: danms(a)us.ibm.com