 Well, hi, thanks for coming here This is a talk for CIS administrators. I don't know how many CIS administrators are here. Can you raise your hand, please? Yeah, you don't have too many free software projects for you, right? Most of them are for developers. So hopefully You like this one Okay What is modular IT? Well, it's really how to explain as everything in you know oops, well modular IT is Distributed we utilize modular integrated and managed service oriented architecture, which is really complex It and it doesn't say much. Let's go through each one of these concepts. Okay Whoops, okay First one is distributed so you can use it for a ministry and manage big deployments of servers We come from the Kinerea Islands. We are islands so we work on one not on land, right? So we need something that Something to manage distribute Systems of servers no matter that no matter how good how we group them or How they interact among them so modularity has that in common as as a requirement It's virtualize if you are familiar with that concept We've been working with it since you know many years ago and you can see here Up there you got the general structure of a modularity server where you have the the server there as a part in system Then we put some manage management tools on there on the party system You got there one of the typical services, which is a firewall is not virtualized for security reasons But all the rest services of the architecture are virtualized Okay, and you have already management tools also on top of the base of the virtualization in order to do many stuff like you know well regular management stuff and This one of those little boxes there. It's one of the services virtualized services Has this basic composition. We have the operating system Which can be centers, they be on windows, whatever The service is that on top of it and each virtual machine has also some management tools What we call a modular architecture, okay, this is also a regular virtualize Schema we have the base system of every or whatever Virtual machine you got you have this for example one service you have More servers on each machine and then you have for example, this one is classified is all we can be configuring in higher availability By default if you want to and then you got you know, you can just put systems services and change services from one machine to another one like a regular Virtualized system the point is that you can do this kind of stuff Centrally manage in distributed servers, which is a cool stuff You don't have to do it everything by hand or at least if you do have to do it by hand You can you have tons of information to do it really quick really fast the services we include our Integrate with each other there's a long way to do in this in this concrete area But I have here some little examples for example the old app and the backup services are mandatory on this architecture This doesn't mean that the old app has to be the master. Okay, we can we have Customers with this architecture integrated in a Windows architecture and the master Master all that is the active directory or the federal one For example, we have the asterix and the fact services integrated. This is something normal for For you know asterix has this by default the point is that we have integrated in two different virtual machines Okay, it's service goes in a virtual machine separate one the faxes Integrated with the email with one nice feature, which is the incoming faxes go through the email or for example the group Where you can you can use the email through a group where or if you don't want to put the group where you can use the email through the default Web interface we put on it Okay, all of them are integrated with the app It's central manage all the hours here should be in the opposite direction. They're all wrong okay for example, we have You can do manual checks through Nages Everybody knows about that when you administrate system centrally we use Nages basically for Only manual check-ins, okay We have some auto check-in and auto collections scripting stuff and Alert systems all the alert systems that the alert system and onto corrections and most of the intelligence of the architecture happens on the server not on the central in the structure That is a requirement. We have because of the networks in the canary islands used to work really bad so Once you if you lose the connection still Check-ins and all the alert has to be going on So you have to pass the intelligence of the architecture to the service instead of having it centrally because of that and We have all these services We are updating from previous versions of this architecture the firewall asset is an apple to deploy. This is a machine for deploying Windows clients, okay and We are right now working on our audio streaming servers some customer asked ask for it So we're working on it. Of course if you guys Develop whatever application or you want to include any application on this architecture is perfectly possible You just have to follow some rules and you can do what we call modularize a service So you you include it by default so you result many problems like deployments Managing Well, all this kind of stuff that many companies don't want to do or is a tedious activity for them They just Resolve it with this and they can concentrate all the efforts on what they do bus, which is the application itself so well, this is as Regular so architecture. Okay made out of free software. What is not modularity? Modularity is not a Virtualization system itself. We use standard virtualization solutions in this case send We used to use CH route before and before that use a model Linux. We have used also in the past VMware Right now we are working 100% on send and Modularity is not a bunch of appliance you can download and try they interact with each other and interact with a The management system. Okay, so it's not something you don't know and try it's much more than that Okay, well, you should try this. Okay first of all who is for Cisadmin basically If you do I said before if you are developing your own application This will solve you many many many Problems related with tedious activities, right? Deployment is probably the first one you see really fast because this architecture that it is deployed really really really fast I mean you send a technician to the to a customer and you have a Server working in one two hours. Okay with all the services you want If you are a IT provider You can use it. We have already other companies using Our system to make money right and it's perfectly possible and Medium-sized companies are the one that really Get advantages of this stuff. Okay, we have modulality in different servers from different customers different hardware In a small bandwidth environments and multiplat for environments You know in all kind of situations we cannot choose the customers we want so we have to We have to this to define some system that fits in many many many different environments Without having to change all the procedures and having to change the technologies so what we have done is Modulality was born as an internal project we use it in pretty much almost 90% of our customers So what we do was just put a GPU on it and just Make it public that was a couple of years ago almost well one year and a half And now what we are doing is not just put a GPU on it and show it but also let people get into it and collaborate with us So that is why we are here There you have all the links Let me show you one at least. Well, this is This is the The gate We are using right now Miguel is a guy that put the new stuff on the project. Then we have a couple more guys that Basically getting charge of their deployments and and go through what The release manager does and change stuff and then we have a third level bunch of guys that Does the the manuals and things like that the other ones that the ground technicians, okay? the ones that go to the customers and deploy staff and This is the red man we're using and I only want to show you Something really Hello There we go. Tic-tac. Tic-tac. Tic-tac. There we go. I did something wrong. Okay. Yeah, that's There we go. Well, this is a summary a simple summary of what we are here To check the how much control we got. Okay, this most of the criticals are due to backup stuff because When when we we back up not just our system modularity services, but also we got backup many of services from other providers in our customers, so Many times it happens that the backup cannot be done So we have to send an alert to them so we solve that our responsibility when we finish that and You can see here something like two two hundred and fifty servers and Something like one thousand and two hundred virtual machines. So that is the numbers. We're talking about When we you know, this is something we have worked on quite for quite some time so You let us grow which is something we couldn't do the before and So we have been around for more than ten years now in the free software business back in the Canary Islands now in the mainland also So like modularity is the result of all that experience and all that job. I So I you already saw the numbers We now can scale in our business. It was something it's something that we couldn't do before and also the last thing I want to say is that Basically we did this stuff because system administrators in our company needed to have a life and It's something it has to happen. I mean system administrators are burned absolutely burn all over. I mean I only once I know I quite burned So this kind of stuff began with that idea. Okay, we have to make these guys live better and The only the only way to do that is giving them control and information of what is happening in their customers so That's pretty much what I wanted to say if somebody have any questions. Just feel free just Go to the Get into the mailing list and that's it. Okay. We all start to send As I told you before we have been trying some others and we know in a few years We will have to switch. Yeah, open V said or Cal VM is something we are trying because you know Saying is now I won't say I wouldn't say in risk But you know some things have changed. So we have to be prepared It's something we have done in the past and for sure we have to do in the future The point is not get stuck in the technologies. This is more about procedures than our technology. We use Pop it with rocks. We use sand. We use medius. We use money I don't know many many many standard Projects we just configure them In a proper way and just put something in on it to to make life easier. We don't have them down, you know Thanks