I help build and run OpenStack based clusters for Bloomberg
I'm involved in the following OpenStack projects: Nova,Swift,Quantum,Cinder,Ironic,Openstack-manuals,Deployment
OpenStack has become part of the biggest and fastest growing compute platform for
Bloomberg (my employer for > 25 years). I've been with this project since (almost) the very
beginning, joining just as it went from a science experiment to something we might use for
production workloads. I was the first to get it working inside Bloombergs restricted core
networks (i.e. behind a proxy).
I believe in open source and the common good it can deliver such as cost benefits as well (in the case of open source cloud solutions) supporting data sovereignty and self-determination.
In a personal capacity the operators meetups were able to bring the perspective of
organizations running large deployments at scale and over long periods of time to the
development community. As a simple example, when your fleet is holding production assets, supporting production workloads and numbers in the many thousands of servers, a staged release of a new OpenStack release with pauses for testing takes an appreciable amount oftime. We also advocated strongly for things like skip-level upgrades and also made the successful case to STOP deleting the documentation for down-rev versions of OpenStack.
Indirectly I'm also thrilled to see the upstream contributions that my colleagues in our
growing cloud group are making upstream, to OpenStack itself, to Ceph, to Calico and to
other projects.
For example we have managed to get some changes merged to help live migration
performance https://review.opendev.org/c/openstack/python-neutronclient/+/966756
and here is a spec for further improvements : https://review.opendev.org/c/openstack/nova-specs/+/970298
Here is a spec we just submitted for per-project control of QoS (quality of service) in Neutron: https://review.opendev.org/c/openstack/neutron-specs/+/965946
Our cloud is predominantly Ceph based for storage and Bloomberg recently had four staff
members join the core RGW (RADOS gateway - S3 for Ceph) team.
We also have work underway to help Calico scale to larger clusters. Calico is our choice of
Neutron implementation for a pure Layer 3 SDN and distributed firewall.
No prior experience of non-profits per se. I am open to learning. The experience of chairing
the ops meetups IRC meetings and pulling together a number of in-person events probably
has some relevance but is not exactly the same thing.
Open Source faces many challenges of the non-code type such as legislation that places
intolerable barriers on using open solutions, bad actors in the communities, as well as
determined efforts to capture the assets of open source communities within corporate ring-
fences. The board can help ensure that the OpenInfra projects navigate these challenges
successfully, spread the word about the value being delivered by these projects, bring the
various communities together and help the relevant open source communities focus on their projects.
Regroup support around the projects and market areas with a real prospect for growth, real
opportunities and focus on those, for example the political situation with US-based hyper
scale public cloud has reignited interest in data sovereignty and self determination with open source infrastructure solutions. In a different area of concern, certain large proprietary virtualization vendors have shone a spotlight, once again, on how open source can provide cost-benefits and resistance against proprietary vendor lock-in.
Chris has already been nominated by: