 Hello. Hey, Taylor. Hello, Michael. Good morning. It is still morning for you, isn't it? Ah, it's for me. It's noon. So that's not not too bad, to be honest. Yeah, we'll give it a few minutes, everyone. If you can add yourself to the meeting notes and agenda, post it on Zoom. If there's anything you'd like to discuss or have a conversation about, please add to the agenda. All right. So this is the third Monday and CNCF Telcom user group. There's two meetings and twice a month, two meetings a month that we have. The first Monday is at 8 a.m. Pacific time and this meeting is at 7 p.m. standard time or 11 UTC. Posted the meeting notes into Zoom chat. We also have a Slack channel on CNCF Slack, uh, Tug. And please join that if you can. We have a GitHub as well. I'll post that to the chat. There's some, the meeting times and other information are there. All right. So does anyone have any agenda topics they would like to talk about? Hi, hi, hi, Taylor. This is Rabi from Vodafone. Hey, Rabi. I'd like to just bring a question. Now, I'm not sure how for you guys aware of CNT activities. Something we start to look at is if you think about OpenNV, for example, we do have groups in OpenNV and tooling to create an instant of the infrastructure reference implementation, as we call them in CNTT. We're trying to figure out what would be the right way of creating a container-based implementation and have that as a reference implementation that we could make compliant to the CNTD specification. Just a question to the group if they're aware of any tooling or any ways of making that possible. It's a general question just like for getting people's idea. I wrote something in there, Rabi, if you could look and see in the notes. I captured what you're thinking. Yes, thank you. I think there's probably several people on the call right now. I'm looking at like from Lidsy, Hey Bill, and other folks really could talk about tooling. From my understanding, there's going to be a lot of different pieces, so probably thinking multiple parts as far as what could be used, different alternatives for bringing up pieces of you're going to have the application side, and then maybe the platform side would be the simplest, but even with that you could start looking at a lot of different components that would be put together. Then there's probably a lot of options, but maybe some folks have some specific things. This probably is something that I would recommend posting to the Telco Music Group mailing list so that folks that aren't on the call maybe could give some responses and feedback on that. Okay, sound good. Thanks. Yeah, I guess if I can just jump in. This is Bill Mulligan from Lidsy. I know Taylor, you guys in the CNF test bit are currently using CubeSpray right now, where you're considering using CubeOne, so I know that those are both tool tools. They're like available open source right now. Obviously, there's all of other solutions on the market too, but I think it would be good to get other people within these two. What might be good, Robbie, is to have a breakdown that's high enough level for people to get into and talk about some of the different parts. You can say CubeSpray is going to bring up a cluster and a Kubernetes cluster with a few specific parts on that. It's not going to include let's say all the different things that are going to be needed in a maybe a production platform versus just saying the cluster itself. What are all the extensions to Kubernetes that you're going to install? Then there's a bunch of other parts. Thinking of what is the CNTT looking at? It's the whole life cycle. Management and stuff of the cluster would be one part. There's also management of the applications and a bunch of other things. Without looking at how would you design that, I'm stepping back and thinking, Robbie, you're saying how would we even have something to start playing with and trying out? I think is what you're asking versus how do we implement the whole thing since it hasn't been fully designed? Yes, exactly. It's kind of something to start playing with, something to start with. It's important for it to be open source. It's important for it to be an active project that we could contribute to or adding some requirements and trying to mature it or evolve it to become that reference or implementation for CNTT-based implementation. That's basically to cover the operators' use cases that they are interested in. A tool or a project with a GitHub repository to allow us to do that would be an idea. That's a great idea. I think if you had some of the major components that you're looking at for RE2 described in a higher level than we could look at, here's some tools that could fit for those parts. There's going to be a lot of alternatives like the cube one. You can deploy clusters and manage the lifecycle of, say, a worker node and other parts and bringing them in and scaling them out. That may be something that's desired. Someone else may want to try something else for the cluster provisioning or whatever. There's a lot there. Then you can say, here's another component that we need and here's a set of tools that could be used for that. If you start there, then we could probably get a discussion happening on the mailing list and probably Slack as well would be a good place to post and link to a message about that. As far as a GitHub repo, I think if we can get the conversation going, it's probably be interest in various projects until you get something going. It may even be like a CNTT, my thoughts would be like a CNTT org repo, just some type of example test lab or something to put something together might be a good place. That makes sense. Tom and I will put something together and post it in Slack and maybe we start the conversation by the mailing list. That's a good idea. Thanks. Yeah. From the CNF test fed standpoint, I could see doing something or as an example, here's some pieces that could work to, if you're saying someone wanted to implement a platform in a specific way, then here's some pieces that you could put together and do something that follows that specification. I could see that as a way to do it. It's almost towards a reference implementation, but prior to that, saying here's an example of some of the tooling and how we would recommend building it starting from using Cloud Native principles, I wouldn't really want to think this is a reference platform though. I think it's a good compliment to what you're talking about though. I definitely shouldn't chat with you some more about that. Yes, indeed. To be fair, with the CNTT, it is what's called reference implementation, which I don't think this bed belongs in there, but also there's the reference certification where the testing is need to be happening. I think more discussion needed to see how that will align to what we intend to do in CNTT. I agree to your point. I think certainly a lot of discussions need to be taken to understand how they fit within what we're trying to do in CNTT indeed. But I think from the other point of view, if we get the reference implementation understood first or agreed on, that would be a great milestone. Then we can look at the certification and conformance aspect of this. Sounds good. I'm posting the GitHub CNTT link into Zoom chat for folks who are not familiar with that. Robbie, would you mind just saying for folks that may not be familiar on this call, real quick what CNTT is? Yes. CNTT stands for Common and the VI TILCO Task Force. The intention behind that is to provide the consistent implementation and architecture of the infrastructure that runs both VNFs and CNFs. We do have two tracks within CNTT, one targeting the VNF and using the OpenNV ecosystem to create the reference implementation. Also, we have a second track, which is focusing on the cloud native and containerized platforms. Tom Kevlin is in the call, he's the lead of the Reference Architecture 2, which is the Kubernetes-based Reference Architecture. The intent here is to try to replicate what exists in OpenNV but for containerized workload. The intention is to find a way to implement that infrastructure based on the Reference Architecture 2 within CNTT, and also trying to develop these cases and conformance tests to start to certify and verify the infrastructure and with the workload running on top of that. And if you don't know more about it, please feel free to visit the GitHub repo posted by Taylor, and there's a meeting, a schedule that you can participate and look in and dive in and see what's happening around there. Thanks Robbie. I think unless someone has anything else, the other thing that we had is the white paper related to the Telecom User Group white paper, which specifically, I say the white paper, there's actually quite a few that have been being worked on. There's a white paper that's the Cloud Native Thinking for Telecommunications, and I'm going to post that link here. Chapter 1 has been being worked on for many, many months now, and to, I think, I think move some stuff forward on that. Tom had some thoughts about maybe moving to GitHub, posting to the white paper folder on the Tug repo, and then creating some issues and PRs for making some updates on the final items. I think that there's maybe a few of the pieces in the paper on the Google Doc, which I posted to the chat here, that are probably ready to close out. There's a lot of stuff that was kind of replacements on items that were agreed on in earlier parts that I can see, but didn't weren't clicked as resolved. So I need to go through that maybe with you, Tom, and make sure that we've closed out all that part so it doesn't, if we can have it finished. But other than that, I think that was about my only thoughts. Do you want to talk about that for a minute, Tom? Yeah, I've copied it over as is at the moment. As comments and suggestions are closed within Google Docs, we can easily modify the PR. The idea was rather than having lots of different comments and suggestions within a single Google Doc, you can raise issues with regards to a particular sentence statement or whatever paragraph, and then that can be discussed within the issue and then the change made in a specific PR. I just think it would be a cleaner way to manage the document, the version controller. If we think we're at a stage where it's ready enough to be in a version control system, that was a theory behind it. And then similarly, when people refer to what is the definition of cloud native network function, in the same way that everybody just links to the CNCF definition of cloud native, we can have a similar, everyone will link to sections of this document that describes what cloud native network function is. It becomes the definition. So that was the idea. In terms of document management, we found it works quite well using GitHub in CNT and other areas. So we're interested in people's feedback on that and whether or not to see it being a useful way of managing the document going forward. Oh yeah, and I can see Taylor's put the PR link up there, so I don't know what to think. I'm not expecting to encourage that now. Like I said, we'll go through and get it ready. Does anyone have any feedback on that? The white paper or moving to GitHub? It's certainly supported to move moving it to GitHub. It's just I think a bit better to keep control of the text and I think particularly the content which Tom posted so far is actually I think in a stage that GitHub would be the most more appropriate tool to handle it. So I think Taylor, this is Ravi for the purpose of this white paper, the chapter one, the content Tom is proposing. I think there's generally agreement on the content, so it might make sense just to push that and merge it and if there's any change needed, then that will be following the PR process. At the moment the PR is introducing that content, so I think once the content is in there, people can start adding their own PRs to change anything if they feel there's a need to do that. Is that the process we're trying to follow here or? Yeah, get an initial version, whatever that is, and there's been a few updates on the current PR, so maybe there'll be a few more based on what's been accepted, but not clicked, resolved essentially, but otherwise have an initial version and then from there out do PRs for any new changes. Any disagreement with moving to GitHub from Google Doc? I think maybe one positive would be folks who don't have access to Google Docs. There's a lot of companies that are not able to access random organizations or Google orgs, so they wouldn't be able to access and there's a lot of different reasons, so ideally GitHub would give access for more people, so that would be a good thing. But if there's any disagreement, if you give a plus one in the doc, if you agree for that to have access to the meeting notes, please go ahead. Yeah, sorry for a quick question. It's late from Huawei. Is there any rules on this PR thing? Because I don't know maybe how long should we wait for, you know, if there's any comments or PRs happen or we just, I don't know how the schedule goes, because it's like when we are doing this with Google Docs, we have many comments as well. It's just like endless questions in the comments. Yeah, that's a good point and something that's come up in the CNTT before. I mean, I don't own the repository at all, but I'm happy to have a chat with Adam and come up with some really really basic contribution guidelines and that kind of answers that kind of question based on what we do elsewhere, unless there already are some Taylor for sort of generic CNCF repository rules. Yeah, we don't really have any rules right now on that. I would say if it seems like we're getting enough plus one, then it's been more of the overall feel if it seems more towards, yeah, go ahead and merge it. It may be better in the GitHub than it was in the Google Doc, where it started to become so large that it can be a little bit harder to tell where which direction it's leaning. You can have someone that may put a whole lot of content and it seems like it's pushed more one direction, but you have several people that just gave a plus thumbs up or thumbs down. And then the history, once we close out a comment in Google Docs, you don't really see where that is. So the conversations in GitHub for PR specifically, even more than say issues would be good. So maybe we'll get a little bit more of that fill, Lee. And I think the other thing would be even if we close stuff out, let's say over, if we just had, leave it there for a week or whatever it is, someone can come back in and create an APR or we can add to the issue and the same issue and say, I don't think this is ready and here's some reasons or open another one. And reference the old issues and you didn't lose the history. So it's probably not as a big a concern to close it sooner, Lee. I agree that GitHub is a much better management tool. Yeah, I agree. All right. So it looks like we have, I don't really see any negatives on getting something in there. Tom, I'll work with you to maybe get some of the updates where they should have been closed out and resolved in the Google Doc just so those are there and people don't have to go back and create PRs for what was agreed on and then we can get something merged in. That's good. Okay. So I don't think we have anything else here. Probably the only thing that I can think that we may have talked about if Dan was on the call was the CNF conformance and I don't know how much he wants to go into that since it's kind of early. Most of it's been mentioned before but at a high level talking about bronze, silver, gold, potentially for how the breakdown for certifications could be and what we're looking at. So there's some efforts to go along with the high level idea for kind of a plan for what could be testing and probably the next call will maybe get into some of that. The high level idea I guess right now is we're looking at having some tooling that can test the usage of say the Kubernetes APIs and how things are going to be running on the platform validation and then how the applications run themselves and doing some testing along that and behave and the difference here from say efforts on maybe some of the other certifications effort the CNTT is it's focused more on are these things following cloud native principles and are they utilizing stuff that would be expected in Kubernetes and maybe some of the things like metrics and monitoring pieces that you could use different tools like Prometheus and open telemetry and as far as those type of tooling how are you exposing the application platform. These are all initial ideas along to kind of move forward the thoughts for the gold, silver, bronze type ideas that would be potentially a test suite that could run and looking at and looking at ideally being able to run that on the CNFTest bed the the end goal would be the test suite itself would be open source and available for anyone to download and run possibly it can end up having two parts maybe completely separate that looks at the actual NFV i platform and then one that would be targeting stuff that's running on there and would have to see where this has compliments and overlaps with stuff like the NTT for looking at the RA2 and other efforts like what OP and FV does that's about it i'm not going to go into all the details right now because i know that it's up in the air and maybe i'll see where dan is on on that one piece back i think that's it does anyone have anything else they'd like to talk about otherwise i think we can give everyone back the rest of the hour here i just sorry i was on mute i just wanted to say i missed the call last week um so i didn't hear exactly what was talked about the the test suite stuff but it was mentioned in frag last week i think in general i think it could be really really useful input into the so not not the RA2 necessarily but the reference certification that comes after so the the point of RA2 is to have something to test against um and the the reference certification will be not only certifying the workloads um but also the platform and i think the sort of tests and the test suite that you're talking about would be really useful input into that um like you say to test the cloud-natedness of the workloads and and how and how cloud-nated the platform is in terms of you know standardized interfaces and whatnot so yeah i think as a concept i think there's a great idea it would be good to discuss it in more detail yeah absolutely i think that um i've i've pointed this out before the the the cloud-native um definition and what i say a cloud-native NFCI platform would be and app the cloud-native applications that could serve telcom should encompass what um CNTT is trying to do so it's essentially a superset so ideally the testing would say we meet all of the criteria for being cloud-native and then we also meet all the criteria for meeting etsy and 3GPP standards and whatever else it's a superset to whatever someone is wanting to implement and making sure that they meet that first so being able to take that and build on it and say here's the additional functionality that we want to ensure is working sounds like a good approach and that would be great if if if we can um figure out how to make that work okay yeah we're good we haven't kicked off the the reference certification work yet but um i know we're going to be discussing it with the next week so um so i'm not sure when the next tub is is it next week or week after sounds good we should have a something useful to talk about in terms of timelines for that and if if you have some insight on when the next um in-person meetings are going to be that would be really good to share um one of the specifically with this group um if you can add to when you find out or if you can add to the upcoming events and that way we can try to push that around i know that it gets um a lot of this will get distributed to other cncf groups and slack channels okay and the earlier the better for getting folks all around the world of course being able to travel where's where's best to post that obviously we can sort of spread it and put it in put it in the google docs meeting notes we can put it on slack do you want it on the adjacent communities uh but in github as well that would be useful yeah i ideally the sand for specifically folks that want to um work with cntt if it's updated on like an events area in the cntt github um or the wiki the cntt wiki either way and then we could link back to that so the upcoming events in the doc or whatever and and being able to send people around it'd be great okay yeah got it okay all right thanks everyone uh the next um meeting is on monday february third at eight a.m pacific that's uh 1600 utc so that's our later in the day call have a good one thanks youtube thank you youtube bye thanks tana bye