 Okay So we will start. Thank you for all of you to stay to this last session of the day I know it was a very tough day for all of you and And it's it's a pleasure for us to be able to speak in front of you Even if it is the last session and even if just after we left the party So we'll speak about a public cloud computing so the the subject of this session is Claude what Claude what use case and what we have done regarding this deployment of open stack to build a European public clouds in European public cloud So first we start by the weather in Europe. Is it cloudy or is it sunny today? I don't know really but what I in say that it is claudifying a lot really claudifying a lot and Claude what is Simply a good example. It's a small company started one year ago. We have started in September last year and Thanks to three men stakeholders orange telco company talas security and defense company and French governments and they all together invested 225 million euro to build a cloud service provider for yes Infrastructure as a service. It's just impressive to see that they have decided that and they have invested on that Compared to all the monsters the competencies the competitors can see all of us words so we have started this this story in September and our main differentiator is based on two main things open source and security open source because we want to be free I want to insist on that and security because we have to differentiate ourself from the competition and we have to provide very high level of security of privacy of Independence see and this is something our customer asking really we have to deliver the type of services in order to be different from the market so The other aspect of this project is to sustain the a global economy in Europe and There's about 1,000 jobs forecasted To two to in five to seven years in Europe thanks to This type of initiatives just to give some flavors today, Claude what it's about 90 people remember we have started from zero and the 90 people were working in Claude what at about 180 people globally working for club even today so Let's speak about our open source strategy our strategy and our DNA because it's it's totally link So open source strategy why a lot of things have been said just before for security reason because in This code normally there's no backdoor There's a lot of a is able to see to screw this code and to analyze this code So it's very interesting for security reason to build solution based on open source second thing is Co-innovation co-innovation because When you are working with that, you know, there's a game today about patent rolling or things like that It's not possible when you are using code coming from community to be attacked by a company for patents for Aspect the other thing is openness and operability for the customers the customers don't want to be locked in a proprietary solution open source is a is a sort of thing very interesting for them because they know that they will deal with an open Code with an interoperable code and they will not have this locking mechanism they will have to afford with other solution and Cost-cutting operation is of use. There's no license cost no maintenance cost on license So it's a way to decrease the overall Total cost of ownership of your solution so it's something very useful when you are busy building your business model to think about open source About sovereignty when you are a cloud provider like us you want to manage your future You want to be to master your foundation. You want to be sure that there's nothing which is not under your own Responsibility you can change what you want where you want and you can manage your future very properly And you want to be independent independent of this solution and no looking mechanism is also something you want for yourself You don't want to be locked in a proprietary solution So when you are building this solution you are making this type of choice. Obviously when you see our DNA It's exactly what you are proposing as value-added for our customer Reversibility if a customer stay in a cloud watch if you is happy with our services He will stay with us if not he can leave and if we want to leave We will give him back all his data all the workload all what is not is property by the way And this is our commitment. So Reversibility is first thing Transparency you want to audit what we are doing want to know what we are doing with your data with your workload You will give to you all the evidence for you to know what is happening there So warranty I've said that that just before but no legacy is very again very important because We have no Nothing coming from the past. Nothing was we it's a full greenfield approach. We are starting from scratch So we have been able to make all the best choice we can do at that moment It's also the low-cost aspect like said before for airline when we make the choice of the best planes the best engine And the lower recurring cost you can find it's that the aspect of low cost we have We have in front of us. So that's the reason why we have make all this choice and that a localization is something really key for European company and especially today if you can guarantee them that the data will be located in the country they want It's a sort of services, which is a really important compared to the recent security Concern you have read in the paper recently so What about open stack and why betting on open stack one year ago look at this wave No way to resist when you have this wave in front of you The only thing to do is to be inside like a surfer If you want to take the benefit of the wave to go faster to go quicker and to have fun by the way Open stack exactly that. It's one of the biggest community who have never seen It's because this comment is growing is growing every day and is proposing something very professional It's a timely released the release are really on time and with new features each time or everything you are expected is delivered and You have in front of you a very strong Ecosystem you can have all the support you need you can touch the best guy at the right moment when you need this guy So that's the strength of this of this community and even if you are a small company You can have an impact on that. That's exactly what we have done. We were three at the beginning no Really technical guys and as soon as the technical guys were recruiting in Claude Watts They were able to deal with this community to give some new features to give some In to decide some blueprints and to and to discuss with the community about that And this is very interesting because you are part of that you can influence What open stack will deliver tomorrow if you are really? involved in this community and in parallel you have to understand that open stack is totally open and Already you can have hybrid scenario. Thanks to open stack Through the API through everything because all the major players are there all the solution are open stack compatible So it's already something you can have you can do At a business level with all your partners and the partners want to come in this game when to play in this game Because they know that because they know that it is feasible and it's very interesting to see the overall ecosystem Now running around Claude Watts just for that reason. They all want to build solution open stack compatible they all want to Transform their own solution to be open stack compatible and that's the reason why it's it's really speedy at the moment And that's the reason why also We are at the level who will see just So sorry Where are we and what have we done in one year look at this and maintain and Just imagine as we try to to reach the roof of the world and and you have this long Way a long journey in front of you We are not at the top of of the mountain and not as a submit We are somewhere in the best camp at six thousand meters Where all the companies cannot be because you have already to make some effort to be there But we know that there's something else to do to to reach the top of the mountain, but we have We are already delivering services. We are up and running since months We have our overall hardware setup and running and we have open Storage services secure storage services for our customers for months is already there's about 5000 customers each day There's about 200 300 new customers per day coming to use our services in our services We have developed a box What we call a cloud what box equivalent to drop box if you want, but the cloud what box and In order for you to to manage properly your collaborative environment you can invite What you want and you can share with them document in a secure way you can manage a role all these aspects for only the people who have invited Are allowed to read this document and to share the document with you and nobody else again And this is a service very useful to avoid this kind of The of shadow IT you can see in in each company. So these services are really used and we have To do that we have developed all around open stack what was described before Develop or integrate some module to build CRM billing ticketing everything we need to operate and the full DevOps System has been set up. We are in a fully Continuous integration mode. We are developed We are able to drop the codes on a daily basis, which is just impressive just imagine all the steps all the stage from development to production going through integration and the pre-production all that is fully dynamic fully manage dynamically thanks to open stack mainly because again environment are running on open stack and and it's fully automatized we can Play all the tests of All the testing all the stressing all the tests for quality of code or for Security of code all that is is automatized. That's the reason why we are at this level for DevOps aspect globally so tremendous job has been done by our ops team and our engineering team to do that The bottom line is very interesting as well because This dynamic ecosystem of partner either vendor or integrator for internet software vendors or value-added integrators They all are working to push and to transform the solution for their solution to be compatible to open stack Let's remember that's a main referential of the cloud is the mware So sometimes they have to understand the small difference between the mware and open stack We help them to do that. We support them to do that and it's Real Dynamic activity we have around us Coming from all those But the game is not totally finished for for us and we are participating a lot In the community to add value and to and to climb this this mountain So our recent contribution is described here So software defined network with desperately needs a fully software environment for this Dynamic aspect of provisioning including compute software and and storage What we call smart smart placement or intelligent placement again a lot of constraints coming from legislation Yes regulation or contract contractual aspect when customer ask for special aspect of security security policy or Data localization on a given country or things like that or if we want is that are close to the compute for Latency reason or or things like that We can do that and we are doing that and we have proposed something in the in the community to do that properly I do point savanna on a vanna a fantastic demo has been done yesterday on that topic It's a distributed. It's a distribution of addup. We have set up in a full cloud Mods with multi-tonnet mode and and we are able to set up this environment very quickly on the clouds for people Who want to make an analytics on on on this data? So it's we are enabling this type of services thanks to the cloud computing platform Okay, and we have also translated all the documentation of open stack in French. It seems to be something Not that important, but there's about 80 Franco fun countries all over the world So it's a way of disseminating open stack widely and to be able to interpret with other Companies who are doing cloud elsewhere in the world So it's it's also very interesting to see the impact of we can have by by doing this type of thing and we're also pushing some Codes in in storage for eras your code or for your replication again the aim is to increase the overall resilience of the solution and to decrease the total TCO so the cost of ownership So we will we have other participation in for the module and we will push on it as an example again to give Some additional means to our customer for them to deploy easily a global system of virtual machine Which is something that desperately need if they want to manage properly the information system We will increase our participation to this analytics and big data at in order to be Meta enabler of this type of services and we will continue on this deployment aspect by Playing with open stack over open stack in order to be able to deploy a large large type of deployment on open stack, so this is What we are doing to grow in this in this system and to reach this the roof of the world and That's the ambition we have and that's where we are at the moment Now I leave the floor to Raphael who wants to pick about the way of accelerating the road to the cloud Okay, thanks It's not finished with Daniel I'm Raphael Ferreran the CEO of in advance and you are a partner with Cladoat since the beginning we work with them in the open stack area and Thanks to Cladoat to help us to progress in this era and to contribute directly to the project So as you can see choosing open stack to launch your cloud platform is the best choice you can make But how can you reduce your time to market? And and be sure you can be ready for production in a short time So to do that we believe there are four things you absolutely need to think about So this is based on our experience with Cladoat First of all an agile methodology for easier step-by-step progress reference architecture for a clear understanding of needs and associated designs that fit them Automation and continuous integration From the beginning you must build a platform that is upgradable and scalable and To finish all in one integrated solution open stack is just a part of the solution So you need a full stack to enter in the game So the approach we recommend is based on an agile methodology. Why? First of all to avoid the tunnel effect. This is a complex project with a lot of interaction So please iterate and stay focused on your minimum valuable product first continuous delivery every two weeks offer a gift to your user it could be a small features or Something more complex, but Try to offer a gift to your user in a replicable cycle So this methodology is also need to allow for continuous innovation and follows a release cycle of open stack Each six months you have a major release So you have to be able to upgrade easily your platform and you need to think about that at the very beginning So this agile methodology always starts with a discovery workshop where we review the customer needs and the objective of the project once the needs have been identified we created a test environment to ensure that Recommended solution works And once we have qualified the solution we start three type of sprints The design sprits to build all the specification architecture diagram and user stories the deployment sprints in where we build the cloud and the CI platform for automatic testing and The development sprints in where we build all the specific features for the provider in this process We are also in charge of the upstream acceptance to be sure as a service provider You will stay in conformity with the open stack core project so in terms of reference architecture the first things To being a cloud provider Is to identify what you want to do and Who you want to be on this market so communication is about repetition. So this is exactly the same slide that my partner Few minutes ago, but that's not a problem So there is no perfect reference architecture, but there is a solution for every provider and One of the key things to reduce your time to market is to be sure you will be profitable is to start With a reference architecture. So find the design test it. We have some benchmark compare several solutions Estimate your cost is take times several months at least At in advance we benchmark adwares in several years. So we are always testing different technologies So we can stay on top of what type of architecture feed to your specific needs and it's a huge huge huge work to do that and to find the appropriate solution for your specific market So through the years We have been able to develop strong relationship with a variety of partner However, our approach has always been agnostic Which allow us to ensure that the customers needs meet and pre priority and provide the best solution for them With goal of limiting time to market and cost while optimizing the technologies that fits so provider needs multiple things But at the very beginning you need physical deployment tools that support deployment on several hardware vendor rollback capacity rapid massive deployment and assurance of conformity So to do that we develop a tool the name is a e deploy It's an open-source tool. You can find it on our github at the in advance and it's a image based deployment tool to have discover on the on the hardware and to automatically configure and provision all the hardware stack You have several projects in the open stack area to do that But we try to work on the open with the open stack community and triple o and this kind of project to federate and integrate all those initiatives in one main project So you you also need to maintain the platform over time for this nothing replace a Capacity to automatically test your use case against new proposed version So for this reason all of our deployments comes with the full continuous integration chain Which is a per queue stoma extension of the open stack C. I Together with puppet e deploy and Jenkins are the core components Which we use to orchestrate and operate physical deployment and upgrades then Service which are deployed on top of the cloud need not only a configuration management tool, but also an orchestration tool and for that we Choose to standardize with it and puppet As I said previously open stack is just a part of the solution But you need a full integrated stack to launch your service and generate revenues So to do that we develop a complete stack with all the necessary tools in this graph the The raid is the open stack infrastructure Which we all know and this is a foundation for all the solution It's 100% open source and we are fully integrated values component Depending on the custom earnings with the goal of reducing time to market so you can find a e-commerce web portal it could be based on my gentle You find a serum based on a sugar serum you find a ticketing tools You find a beaming tools and we developed a bus to integrate all those components In the same way on top of an open stack project and the open stack installation could could be Red Hat installation or Ubuntu installation or DBN distribution Whatever we try to standardize and to keep the open source approach for this project So just in a 30 minutes not possible to give you a complete overview of cloud deployment like cloud water So the best way to move forward is to have a talk So we have a booth in the expo hall tomorrow. We organize one of the official party in the aqua harmony club and for European people or If you want to come in Paris on December 5, we will organize the French open stack event in Paris So invite you to join us So so so Europe is back. That's what we said We are in Hong Kong today seven thousand something like that. Let's imagine that in November last year We can be ten thousand in Paris Yeah, it's a good challenge. Yeah, it's good. So hope to see you in Paris Merci You have question We have time for question before going to the party No question. Okay. Thanks everybody. Thanks. Thank you