 Hi, thank you for joining us for the Newton design series for Neutron with me. I have Armando Armando. Please tell us a little bit about yourself Hi there. Yes. My name is Armando Miyacho, and I've been in the Neutron PtL since the Mitaka release I work for a Hewlett-Bugger Enterprise and I'm based in the San Francisco Bay Area working for RobinStack and then OpenStack for a few years now Very nice. And can you tell us a little bit about Neutron as a project as well? So, yeah, I mean a few words and Neutron is the networking project within the OpenStack ecosystem its mission is to create technology agnostic abstractions to address the networking needs of service providers enterprises and telecom operators that deploy OpenStack requirements and Yeah, we basically focus throughout the entire ISOC stack, you know, from out to all the way to L7 services Got it. So we're about a month approximately from the Austin Design Summit So what were some of the hot topics that you and your team discussed at the summit and what were some of the outcomes? You know, we have a number of you know exciting sessions There's usually lots of yelling But yeah, I mean, it's good discussions and we've agreed that a number of key areas and on the path forward for For a few things that I think we're along overdue and in particular we talked about how we could transition the existing Neutron clients over to the OpenStack clients Which is, you know, this the latter is, you know, something that's been going on for a while too and you know we Talked a little bit about how again, how to make that transition happen and other we also discussed on how how to provide More sophisticated troubleshooting tools for users and operators. Again, this is also like an overdue area of improvement And we also talked about how to streamline our our internals You know, software management processes So that the various initiatives that happen under the Neutron project Umbrella or as you know, so-called stadium get better visibility within the wider ecosystem And you know, lots more Got it. And so The primary audience viewing this design series interview will most likely be operators and users So could you highlight a little bit in Newton design in Newton What you plan on doing that you identified as a user need that you're addressing Yeah So why are that we're going to invest quite a bit this cycle is How to improve the ability to troubleshoot networking issues? I mean We talked about this for a while now and was really converging to a solid abstraction I mean, and this is particularly important for for users and operators who are faced Usually with that, you know with the with the ability to effectively troubleshoot and pinpoint where where issues exists and Helping operators dealing with deployment scalability issues is also another key aspect that we've been discussed extensively Especially in relation to to the software routing solution that Neutron provides Modeling new ways to deploy Neutron over existing data center networks is also going to be one of the major as human that we hope to make Happen for new team Neutron and there's many more Great sounds exact exciting and agree. I think that's definitely user need that needs to be addressed So overall, what are the top three priorities for new features or enhancements to existing ones for the cycle? Okay So you I think It's difficult to come up with like just three but I mean you guys expect to see three major enhancements at the end of Newton One is better troubleshooting capabilities as I've said a few times now And this is beneficial like for a for a full stack orders, right? Anyone is going to be happy to see that another key achievement that we're aiming for And particularly that which is and if he focused he is about providing the ability to trunk Many net to Neutron networks together and expose them to a virtual function be a single attachment point This is this is also going to become relevant for for better metal deployments like ironic and I think finally we've been working on this will be working this for like for a few for sometimes a very fusile now and you're going to Provide a new way to model data center networks, you know an L3 centric fashion She's a big shift from from the way in you turn has been operating So yeah, this to me I see like the top three priorities as well as challenges, but yeah, hopefully you know We won't we won't disappoint our you know our user base You got it and now if we go up one level if you will so the product work group and offset community We've been using this concept of themes at a broad, you know at a broader level to kind of certify buckets of focus for projects So the themes that that exist our scalability resiliency Majubility modularity and drop ability which teams that do you think you turn is focused on for for this cycle? well, we try to You know strike a balance and and all of these things together So, you know, do you see that there are things within, you know, the larger group of folks who are, you know, focused in Newton We're gonna spend time and and an energy in in tackling issues in each of these like buckets if you will and As far as scalability goes, I think the primary areas that that The primary areas that Newton delivers are like in L2 and L3 services and We have been working to identify and resolve like control plane bottlenecks for both L2 and L3 operations Now, you know, we reached a point where Newton is deployed to scale and you know We're getting huge feedback as that some things, you know, don't you know, don't work as they You know, they should be and there's been an ongoing effort in that addressing how, you know Control planes scales out when dealing with the creation and management of of certain L2 and L3 constructs that Newton handles As far as the resiliency goes Since since again the possible it is now possible to create some of the routers that are both like distributed and highly available and In Newton, I think we need to we need a few extra iterations to make that capability rock solid and gain behave well at scale You know thinking thinking about manageability Back in Mirage, we will prove the experience for the end user greatly with With the orchestration, you know, with the get me a network orchestration As it's known That capability allows users to provision network topologies at the push of a single button Like a simple API and the Newton we want to iterate so that That provisioning model can become more flexible and address the various networking topology needs that our readers may have So you see again an increase, you know, better better manageability on that front I Think you mentioned what the lettery to do, right? and the journey of breaking down the new term code base in manageable basis has started a few cycles back and The journey is obviously still ongoing. I mean, it's this loss that still need to be done And we were learning one book sense and you know what doesn't and you know, it's it's a it's a learning expert process and an experience so we do learn through experience and Again, I mentioned the stadium the Neutron stadium before which is sitting is namely the elected the list of projects Initiatives that the new project initiatives that the Newton team and Dorsey's and as and that has going there's been going through a number of transformations and and The transformations, you know, we're a me that addressing this very need of modularity How, you know, how we can, you know, we can improve the way the various beats the pieces fit together And one aspect that we are gonna be focusing on amongst others is how we can improve How the load balancing as I said this module within Neutron fits within, you know, fits between Newton and resolve some of the integration issues that exist To deliver like to deliver load balancing services effectively So as far as the interoperability goes one area that has been discussed for a while now is How to effectively connect like to the link across, you know, two or more separate open stack deployments So it's not just about like federating like identity, but it's also how can you make like two separate networks behave like if there were one and there have been a number of proposals on the table for a few years now and I think that we got your point where there's no critical mass and I think now is the time is right for looking at these The side-by-side and start defining, you know, which one makes sense for which use case because not, you know, that we want obviously You don't ever see the bullet that it's like one size fits all so Again, we have some work to do in order to better identify how, you know, how a solution is fit for For what for type of use case and we aim at, you know, the aspect too And yeah, I mean that would keep us busy for a while for sure Absolutely. Well, thank you so much for joining us Armando and covering Neutron before we go Is there anything else you'd like to add for our viewers? No, I mean, usually I've been asked, you know To, you know, what happens after after the you know, the the current release So what what's like what happens in ten twelve months time? And I usually I usually struggle to give an answer to this this type of question because you know, when sources you're never really in full control and We usually are very ambitious. We try to end up, you know, we end up Over-subscribing ourselves and I see some of, you know, some of the things that we talked about during this call and some of the things that we wanted to You know, we want to tackle in Newton will end up obviously, you know, spilling over Ocata and And typically, you know, some of some of the efforts that I even mentioned like during this call have started prior to Newton So they are typically like multi-cycle efforts and Hopefully, you know, we'll Throughout this journey, we'll still be able to deliver value at the end of each release. I'm probably stating the obvious But it's always good to like keep being in it because otherwise Again, if you're like, okay, at the end of the release, whatever actually done Yeah, I think it's important to iterate that it's multiple iterations. It's a lot, you know, and ongoing journey Awesome. Well, thank you so much Thank you. Thank you for having me You