The day one Keynote started with a great welcome and with some numbers about this Japanese edition: actually at the event there are 5.000 people from 56 countries. That’s the largest Openstack event we have seen outside of North America!
Jonathn Bryce, the Executive Director of the Openstack Foundation, went on the stage and started talking about Openstack as a very global community; he remembers the 3 gold members in Japan: Nec, Fujitsu and Hitachi.
As a community, the project always needs for talents, so 2 years ago there was an initiative from the foundation to create a training marketplace where training providers could list out courses and sessions to help all the experts. Now Jonathan has been happy to announce the 1st certification for individuals “Certified Openstack Administrator”. He mentioned that they have been working with universities and now they are launching a professional certification in order to help put programs together and help establish a good standard baseline of expertise and skills. The course is going to be delivered virtually and the tests will start in 2016.
He started then talking about the recent announcement of the 12th release “Liberty”. It’s the biggest release yet and 2.000 developers contributed to it. The new aspects regard 3 important themes:
- Manageability
- Scalability
- Extensibility
The first two features are very important in order to improve the user experience. As Jonathan says, any time you have a software, you have 2 groups of adopters: developers and users. The core point of the new Openstack development model is “Everyone can be involved in every face of the process”, and users have the opportunity to be in the middle of it. Jonathan states that they have implemented a “working group” mechanism to allow users to have a voice. In this way they’ll have the ability to get involved and to impact the direction of a software and to contribute directly to it.
Talking about “Extensibility”, the Liberty release is the 1st under the new project organisation model. The goal of it is to recognise and manage, to encourage more innovation and more projects.
Bryce made than a clarification about Openstack and the connection with all related projects: looking at Openstack and at data they get from users, we can see that there are some basic services deployed the 80-90% of the time; they cross every Openstack environments and they are the base functionalities of compute storage & networking. Then we can see great functionalities, great innovative technologies (like Big data analytics, relational databases services, etc.) that provide critical business functions. People in organisations can optionally deploy also these functionalities if they need the related specific capabilities.
So these are the bases of this really powerful model: we have a common baseline plus all these additional capabilities built around it. And the common baseline is nothing else then what we talked about in Vancouver: the interoperability that grants a great user experience to all Openstack operators.
To talk a little bit more about interoperability Jonathan introduced Egle Sigler, principal Architect at Rackspace. She started saying that interoperability is really important because it provides a common user experience to users, without “special flavours” of Openstack. Then she went in depth with Defcore, a community development process, started from a capabilities score process; then they brought this score back to the community for a feedback, starting a process of user engagement.
She suggested than 2 ways for people to be involved in pushing Interoperability and Defcore forward: the first is providing data through Rackspace; the second is participating to Defcore meetings.
After Jonathan came back on the stage and, remembering the innovation quadrant he talked about in Vancouver, started talking about adoption and maturity of Openstack solutions: looking at 2010, when Openstack was starting, Nova compute service was in development; and there were 2 organisations around Openstack: Rackspace and Nasa. In 5 years Nova and Virtualization moved from an experimental phase to being winner technologies widely adopted. Following this scheme, looking at different Openstack projects we can see that in few years a lot of projects have climbed from experimentation to wide adoption.
There was then the announcement of a new website section (Project Navigator) where people can find informations on functionalities and innovation of Openstack ecosystem and projects. This system consists in taking hundred of thousands of data from user surveys, from technical community tagging, etc., in order to deliver informations like:
- How we deployed a functionality
- How long did it take to deploy it
He remembered that it’s an initial release, so any feedbacks will be appreciated. Using this topic, he introduced another interesting thing that have been developed. He mentioned some new developed technologies like Sahara (Hadoop and Big data analysis) and Magnum (manage of containers) that had great success during the past years and found the powerful common element: all have been developed in a community, so they are an all in one platforms. In the same way, looking at emerging technologies, like Docker, Mesos and Kubernetes, people want all these to work together.
Then Jonathan introduced Lachlan Evenson, Cloud Platform Engineering Team Lead @ Lithium Technologies that wanted to talk about his journey with containers to production on Openstack.
He explained that at Lithium they help brands to connect, engage and understand the customers through online communities and social monitoring tools.
He showed also some apps:
- Realtime: it runs completely on containers on Openstack. It aggregates globally all customers data and put them on a map in order to show what’s going on in the community in real time.
- Monitoring and logging: he showed dashboard with an overview on “how resources are being consumed”. He underlined that they have 122 running containers in production across 68 different images.
At the end he took a DEMO from commit to deploy containers in less then a minute on Openstack and Aws.
With his intervention he wanted to underline how we can use the Openstack infrastructure, like containers orchestration, with no engineering effort and no more capex expend.
After that, Jonathan introduced a user that is making use of different technologies and integrates them into Openstack: Takuya Ito, Sr. Manager of Infrastructure Engineering and OpenStack Blackbelt at Yahoo! JAPAN.
Takuya started presenting some extraordinary numbers about the company:
Yahoo! Japan has 65 billions of Page views per month; only the site for smartphones has 32 billions of page views per month. Over 270 million Apps have been downloaded; it has plus than 100 services like news, shopping, navigation, movies.
A statistic of Yahoo JP operations on only Openstack environments counts more than 50.000 sunning instances; 20 PB in the data storage are dedicated to Openstack environment; they have more than 20 Openstack cluster running.
Yahoo JP Openstack environment has grown more than 100% in one year in instances, machines and racks.
One of the most interesting report shows that Yahoo JP has moments of spikes caused by national disasters. In fact people go to their website and applications in order to understand what it’s happening. For this specific reason they have some Apps to give informations to people, like:
- Urgent notifications for disasters preventions (earthquakes, tsunami, volcanos, etc.)
- Weather report
They use some of these mission critical apps on Openstack. As an operator of the datacentre, they need in fact to provide resources rapidly, in particular in case of emergency. It’s important to have some APIs in any operating environments.
Their mission is to make an abstraction of the datacenter and Openstack is the core point.
Before 2013 there were more than 10.000 instances running on in-house built IaaS, and there were unique APIs. After, with Openstack, there was a development of an Opensource environment with features like:
- Common API available
- Appliance functions developed with vendors
They wanted an evolved datacenter (with the Datacenter Lifercircle Management) and a datacenter Abstraction. Moreover they reduced the costs.
They consider the co-creation very important, and Openstack is an infrastructure that support people in this.
Then Jonathan took the stage and re-iterate few important points in the concept of “One platform with all the other capabilities: Virtual Machines, Bare metal and containers”.
At the end there was the announcement, by Mark Muehl and Shilla Saebi from Comcast, of the Super User Award! The 4 finalists were: FICO, GoDaddy, Lithium and NTT Group and the final winner was NTT Group, the largest Telco in the world! Our compliments to them!
Our Day 1 blog post ends here. Stay tuned on our Twitter account for any update on this fantastic Tokyo Openstack Summit!
Click here to see the video of Openstack 2015 Tokyo Day 1 Keynote: https://www.youtube.com/watch?v=Prm2_E3ONEw
