I contributed a driver for NFS backup service of cinder volumes for kilo as well as a number of fixes and enhancements to the NetApp cinder drivers. In Liberty, I wrote a blueprint spec that proposed decoupling the cinder volume and backup services, such that they can run on separate nodes, and such that multiple backup service processes can run concurrently - along with POC code for same. Also, implemented QoS specs and scheduler-based over-provisioning for the NetApp drivers, as well as helping out with Manila from time to time.

In Mitaka, I want to continue with the scalable backup work and pursue overlaps with current work to provide active-active HA for cinder.

Finally, I have a long background in networking before moving over to storage and am keeping my eyes out for ways to leverage that as well.