VDSM Potential Features

Revision as of 21:00, 18 June 2012

• Simplifying using QEMU/KVM
  • consuming qemu via command line
    • can we manage/support developers launching qemu directly
  • consuming qemu via libvirt
    • can we integrate with systems that are already using libvirt

• Addressing issues with libvirt
  • Are there kvm specific features we can exploit that libvirt doesn't?

• Scale-up/fail-over
  • can we support a single vdsm node, but allow for building up clusters/groups without bringing in something like ovirt-engine
  • can we look at decentralized fail-over for reliability without a central mgmt server?

• pluggability
  • can we support an API that allows for third-party plugins to support new features or changes in implementation?

• kvm tool integration into the API
  • there are lots of different kvm virt tools for various tasks and they are all stand-alone tools. Can we integrate their use into the node level API. Think libguestfs, virt-install, p2v/v2v tooling. All of these are available, but there isn't an easy way to use this tools through an API.

• Host management operations
  • vdsm already does some host level configuration (see networking e.g.) it would be good to think about extending the API to cover other areas of configuration and updates
  • hardware enumeration
  • driver level information
  • storage configuration (we've got a bit of a discussion going around libstoragemgmt here)

• performance monitoring/debugging
  • is the host collecting enough information to do debug/perf analysis
  • can we support specific configurations of a host that optimize for specific workloads
  • and can we do this in the API such that third-parties can supply and maintain specific workload configurations