 Amdox is a leader in providing software and services to some key industries like telecommunications, media, and financial services. In our next session, we welcome Cedric Jegu, who's the head of technical product at Amdox, and we'll learn about Amdox modernization journey and how it added value for their end customers. Cedric, welcome. Welcome, welcome. Good, thank you. So describe your modern application, your portfolio, and what you're delivering for customers. So Amdox is a BSS OSS player, so we are providing a full digital suite for customers. Our customers are communication service providers, which are half to deploy a full digital suite, customer experience, for the full BSS OSS stacks. So actually Amdox is one of the leaders in this kind of digital transformation. Yeah, so of course, you talk about OSS and BSS, I mean, you're talking about some really hardened stacks, right, the telco industry. Say what you want about it, but boy, the phone works when you dial it. So you've got sort of a decades old platform that you guys have been evolving over the years. Describe this modernization journey and the role that CouchBase played. What value does this offer, this modernization offer to your organization, and where does CouchBase fit? Yeah, exactly, it's the same so that, so basically what all solution is, it's a portfolio of a large number of components which are to deal with from the experience of the user and from, and then dealing all the activation of the services in the network in order to deliver a solution, a pure service like a mobile services or communication services to the end users. So we have a full suite, which was previously based on technologies based on the Oracle with WebLogic and things like that. And what we did is that we do a modernization of this something like six years ago, a bit more than six years ago, we start the modernization and transformation of our product into a cloud native solution, cloud native solution. So, and when we did that, we start with CouchBase as a partner to provide the cloud native database. So we are actually delivering, we have an R&D of more than 8,000 people developing this product, it's a product which is used by more than 300 customers. So it's real product that needs to be very flexible, that needs to address many kinds of use cases from our telco or customers, which are CSPs, usually tier zero, tier one telcos. So we, what we wanted to build is a full cloud native solution that can work on any cloud, then can scale very easily and can address multiple use cases, okay? And that's why CouchBase, when we selected CouchBase, it matched a lot of requirements and criterias we had. And then when we decided to modernize our product, we decided to work with CouchBase. Yeah, so you had a lot of experience and legacy with Oracle and WebLogic. I'm curious to sort of follow up. Why didn't you stay with Oracle? You mentioned, got to run on any cloud, got to be flexible, but could you double click on what CouchBase delivered from a requirement standpoint that was such a good fit? Well, the good fit with technology which has CouchBase is first, it's a NoSQL database, right? So it's in terms of performance for some of the use cases that we have. It's very important to have technology which are ardent and optimized for the NoSQL use cases, that's the first thing. The second thing, as I mentioned, the scalability. The fact that you can almost infinitely, you can increase the size of your cluster. You can add the more servers and this will scale very rapidly. And also what were very interesting to have from CouchBase is the ability to have something which can be replicated across multiple sites. So with the XDCR technology from CouchBase which enabled to build very modern architecture with deployment on multiple regions to have disaster recovery, active sites, things like that which are becoming like the main requirement for more customers now. Okay, so I'm presuming there were parts of your application portfolio that you weren't going to touch and throw away. You had to connect the new with the old. That's always a challenge. I'm wondering what advice you'd give to an organization that's kind of investing in a similar path, trying to deliver the best digital experiences to customers. What would you say are the modernization you got to have, must-haves, whether it's architecture, internal culture, what are some of those items? So Dave, yes, you're right. I think the integration with the legacy systems is actually a very, very important topic in our domain, in the telco domain. But we made a very, I would say, drastic choice or brave choice when in six years from now, when we decided to reformat, to replatform, sorry, completely our portfolio. Okay, so we changed more than 95% of our portfolio and 95% of our portfolio today are cloud native which means that they can be deployed on any cloud that actually they are fully scannable. And still, we did this transformation now when we do the digital transformation of the telco of the customer system, then we need to integrate with legacy systems and we need to help our customers to migrate from their legacy systems to cloud native solutions. And doing so, it's important to have in the database domain, it's very important to have a solution which is very flexible in terms of what kind of data it can manage. And I can, as I said, scale easily for sure, but also it's secure, okay? Because when you are moving the data from a legacy system or a code base or whatever to another type of database, you want to be sure that you can do it securely and you are not compromising in any sense in terms of security, scalability, et cetera. So in this case, I mean, I will say and in this approach, in this journey, code base was very, very important component in our strategy. For all the reasons I mentioned, right? It's very cloud native, it's scalable, it's secure. It's another product telco grade. So that's why it's really the key aspect here. You know, this notion that 90% you really replatformed 90% of your portfolio and made a cloud native, that's a brave move because a lot of companies do that I've talked to, they'll build an abstraction layer and microservices and make that piece cloud native and then have that kind of overlay. You decided not to do that. Why is that, was that for performance reasons? You were worried about bringing along technical debt. I mean, that really must have been an interesting discussion internally in your company. Yeah, Dave, it's true. I mean, the main motivation, the main driver was business flexibility because now we live in a world where our customers, what they need is to be able to test a new feature quickly. They need to be able to scale their system in a matter of hours, okay? So we are not in a domain anymore where when you have to upgrade something you need to take a few days. It needs to be done very quickly. And the only way to achieve those requirements, these business requirements is to be cloud native. It's to build microservices and to really rely 100% of microservices architectures because this is the only way you will have the business flexibility you will be able to have a resilient architecture. You can deploy this with a full high availability across multiple zones, multiple regions and things like that. So any modern architecture today that is competing with us are actually a microservices-based architecture. There is no other way to achieve, to meet the requirement of the market today. And especially when 5G is coming, things will become much more complex, will become much more distributed. You cannot work anymore with a monolith architecture. And again, I think the database is no way different. Needs to follow the same kind of architecture. Needs to follow the same principles. So that's why I'm having another point about Roger. Yeah, so if I summarize, it sounds like your top three requirements would be flexibility, which you're getting from the cloud native and microservices piece, the scale and the security. Is that right? Did I get that right? The three top requirements? That's right on the resiliency as well. I mean, the fact that now with microservices architecture, if one of the system is done, he knows how to restart itself, right? Itself, sorry. So that's this kind of architecture that we built. It's an architecture which can be resilient in a sense that it can sense itself and it can ensure full availability. And if something is going down or it's not working properly, then all sorts of mechanisms will happen in order to go back to a stable state. Yeah, so you've got that automation in there so you don't, doesn't require the labor that it might have 10 years ago. So you're obviously embracing cloud native, microservices. So you're on that journey. I'm curious, what are you doing with that? You're free, you're freeing up. You guys used to bring in lab coats and dig in and figure out what's wrong or restart the system. Where are you in your journey and how are you sort of reallocating those resources and where do you see that going? Yeah, okay. So that's a very good point because actually when we build this new system which is unable to self heal himself, right? Actually the question was more about how we can improve the system even more. How we can be sure that issues that any issues which we are facing will not happen again, will not occur again, okay? And this is an SRE principle, okay? Practice that we have, now people are working on automation. They are building automation around all of these recovery procedures or about fixing. So they are not actually digging into the application now anymore into the system. They learn how the system is working and building all the right automation task to ensure that the system is constantly resilient, right? So that's the SRE practices. Our organization is now built around this kind of this approach, DevOps, pure DevOps being fully agile obviously, having a SRE organization or SRE oriented organization. And that's the only way you can reach very high SLA in terms of availability. So the big problem that your traditional telco customers have is the amount of data that they're servicing going through the roof and the cost per bit is sinking like that. And you have all the over the top providers coming in, creating these customer experiences with modern applications and they've owned the customer data. You mentioned 5G. So I'm interested what the future of modern apps looks like for Amdox and your customers because 5G gives your traditional telco customers the ability, if they can have these flexible systems that you're providing to now have better relationships with customers and actually kind of reclaim some of that value that they've lost to a lot of competitors. Your thoughts on the future. So first, technically speaking we will have two challenges. One is about data and the other one is about distribution of the workload. Because when we are speaking about 5G we are speaking about the age, we are speaking about the fact that an application may be located very closely to the network because it needs to be to achieve, to deliver very short latency. And this application can move. So you will have to be able to distribute completely your solutions. And that's why we're working closely with the cloud providers at AWS, Azure, Google. And because we need to be sure that the applications or the systems that we are building will be able to distribute the application as close as possible to the own users. So that's one of the key challenges. Which means that the application needs to be very portable, needs to be very scalable. And then it needs to be able to move very quickly from one place to another. That's really what is happening now and what will become the norm with 5G. The other challenge is behind the communication of all these components is really the data because now we will capture more and more data coming from the different systems. And I'm not speaking only about the customer data, who they are, what they like and what they want to do, et cetera. I'm speaking also about the monitoring data of the systems. So we will generate a lot of information and this information needs to be treated very quickly, needs to be stored in very large data lake. And we will need to have extraction and manipulation of the data very quickly to give the right informations to the applications. In this case, it's very important to have application to have databases that can, as I said, scale very quickly but also will be able to have very high-density node, sense that with a certain amount of memory or sentiment of storage, you can store a lot of data. And this is where we are always checking what is the best technologies and so forth. Coach-based is the technology that we're using for storing all those data because it's the ratio in terms of performance on the number of data you can store is very high. So that's another challenge that we're addressing. Of course, Coach-based is not the only solution but it's another one. Excellent, okay. We've got to leave it there, Cedric. Thanks so much, a great story and really appreciate your insights. You're welcome, Dave. Thank you very much. Okay, that's it for today. I hope you've enjoyed the Application Modernization Summit made possible by CouchBase. We shared some fresh survey data and got the perspectives of three expert analysts. We got an outstanding roadmap from Ravi Mayaram who's the CTO of CouchBase. And of course, we got the customer angle from Cedric. So look, maybe you're an organization going through a modernization initiative. And if you're thinking about what the future of applications looks like, check out CouchBase on the Road this summer, the Application Modernization Summit, it's hitting the road, traversing North America and Europe. Find out where there'll be near you by visiting couchbase.com slash roadshow. Ravi is going to be there along with other thought leaders and peers who will be sharing learnings and best practices on how to modernize now and for the future. And you'll get a chance to interact with some of those peers, something that everyone I know is looking forward to. This is Dave Vellante. Thanks for joining us today and thanks for watching theCUBE.