It’s time for the community to determine the winner of the Superuser Award to be presented at the OpenStack Austin Summit. Based on the nominations received, the Superuser Editorial Advisory Board conducted the first round of judging and narrowed the pool to four finalists.
Now, it’s your turn.
The team from Beftair is one of the four finalists. Review the nomination criteria below, check out theother nominees and cast your vote before the deadline, Friday, April 8 at 11:59 p.m. Pacific Daylight Time . Voting is limited to one ballot per person.
How has OpenStack transformed your business?
It’s mainly linked to a cultural change. We’ve gone from ops and dev to having the dev teams capable of running their own estate and supporting it. We’ve gone for EVERYTHING-as-code so devs simply check-in JSON files to control their Platform, OS, Disk mounts, Network, LB, Code…and we’ve put it all in pipelines. We’ve also totally used the same approach for the infrastructure – so all our Openstack, hypervisors, firewalls and routers are all code controlled with their own pipelines. As a result of this we will go completely immutable – every release will rebuild from scratch the machines, code, storage, networking, vips and green-blue deploy. We do around 500 releases a week – we’re trying to accelerate this to 500 a day….
How has the organization participated in or contributed to the OpenStack community?
We’ve suggested and committed code/features upstream. We’ve worked with communities (specifically around Ansible and Nuage) to provide publicly available modules to help others automate this relatively new area. We’re pushing a ‘reference’ stack with hand chosen vendors and tools – both hardware and tooling. We’ve hosted events – taken speakers to places like Romania (where the OS community is fledgling) and will be speaking at both the OpenStack summit and the Red Hat summit this year as well as several other London events. It’s all on
What is the scale of your OpenStack deployment?
It’s big-ish We are working towards:
· 1,300 hypervisors across two data centers. I think we will end up being Red Hat’s third largest installation worldwide.
· This will support: ~200 applications serving +120 million daily transactions, +2.7 billion daily API calls. We monitor 120,000 data points a second, we ship 2.5 terabyte event logs daily.
· We will scale up form 500 deploys a week to 500 a day.
How is the team innovating with OpenStack?
I don’t think anyone has done the full everything-as-code approach at this scale and is prepared to be so public about it. The SDN is new and is the biggest innovation – to allow devs to spin up their own networks – and to deploy network changes not only through a pipeline, but also green-blue style is pretty rare. Our move to a complete immutable deployment model is our biggest internal change and some of the innovation to handle this for an exchange running at such scale is quite technically impressive. Soon we will begin our bare-metal work. This will use exactly the same delivery tooling but spin up a bare machine. This will allow us to cater for some of the exotic components we have as well as making containerization (mainly for our tooling) even easier. Of all the places I have ever worked (and my guys say the same) never have I been given such free reign to design the BEST solution for performance and efficiency. Betfair has been absolutely unique in this endeavor and is setting the scene for those to follow. The icing on the cake is that we can talk about it – as such we are playing more and more a role in external events and (of course it’s a great recruitment tool) I hope pushing the London tech community. I hope we will be as prolific as Etsy in this manner…
Cover image courtesy Betfair.