Speaker 1: 0:26

This is the Art of Network Engineering podcast.

Speaker 2: 0:37

In this podcast we'll explore tools, technologies and technology keeping.

Speaker 3: 0:41

We aim to bring you information that will expand your skill sense and toolbox and share the stories of fellow network engineers. Welcome to the Art of Network Engineering. I am Tim Bertino at Timbertino on Twitter and I am joined this evening by Lexie Cooper at Trackit Pacer on Twitter. What's happening, lexie?

Speaker 4: 1:03

Hey Sam, not a whole lot going on. It was just the Fourth of July recently, so everything is like chill and very hot and I am chill and I don't have AC in my house, so that's fun. Yeah, that's where, that's where my brain is no AC.

Speaker 5: 1:23

Oh, my yeah.

Speaker 4: 1:25

Well, okay, technically, one room in my house has like a window unit, so it's been fun in Seattle summer. I don't know, I don't think you.

Speaker 3: 1:34

I don't think you did it on purpose, but when you said it was the Fourth of July and everything's chill, you kind of put your hands in front of the camera and I just saw that, as you're talking about the Fourth of July and we were making sure you had all 10 fingers, yeah, you know what?

Speaker 4: 1:48

though. You know what, though? I definitely did not have any danger this Fourth of July, because my neighborhood, instead of doing fireworks, did a drone show, which to me sounded super lame, but well I thought it'd be super lame, but it wasn't.

Speaker 4: 2:02

It wasn't. It was actually really really neat and I like it better than fireworks now, because it was like quite like it wasn't the scary fireworks noises. But it was also, like you know, they had like a soundtrack. You know, like I don't know, it was like 80s rock songs. This was so cool. They did like 3D, like animations of, like you know, the Liberty Bell and the Statue of Liberty.

Speaker 4: 2:21

Really, there was like an eagle that flapped its way. It was awesome, it was very cool. So I highly recommend drone shows for your special occasions. How was that? How was your holiday, tim?

Speaker 3: 2:34

Great. We spin it on my wife's parents' farm and they've got the whole thing of. So my father-in-law used to work for the city in this small town and at one point years ago he literally came home. The city or the town playground, got all new playground equipment. So he brought home one of those like tall sheet metal death trap slides that we all remember from the 80s and 90s yeah, hooked a hose up to it and in their backyard that slopes it like a perfect angle downhill and there's, like this, this big water slide in there in their backyard on the farm. So that's been like a fourth of July tradition for the last I don't know 10 plus years.

Speaker 4: 3:26

That's awesome. That sounds like a great time, oh, kind of jealous, like it's been a while since I've done water sports, even even slip and slide, type things. That's cool.

Speaker 3: 3:37

And I don't. I don't get too much into the fireworks anymore, which makes things a ton safer. So that's good.

Speaker 4: 3:45

It's good and less scary for children and dogs.

Speaker 3: 3:49

And dogs? Yep, exactly. So we did already have a question in the chat from Zatharian, but are you on threads? That's a new thing now.

Speaker 4: 4:00

Oh my gosh, the next Twitter competitor.

Speaker 3: 4:04

I haven't. I haven't had a chance to talk to you about it yet, Lexi. What are your thoughts? I know?

Speaker 4: 4:08

Oh my god.

Speaker 3: 4:08

I don't do it.

Speaker 2: 4:09

I'm so tired, I'm so tired.

Speaker 4: 4:15

I'm so tired. So I'm on everything. I'm on everything except for threads and Snapchat and Facebook. Okay, those are the three things, but I am on Blue Sky, I am on Mastodon, I guess the, and I'm on the actual Twitter, right. So, like you know, an Instagram and YouTube and Twitch and all the blah, blah, blah. And I am so tired of opening up accounts and then reposting because there's not a single program that can. Everybody's closing down their stupid API is right. So no, everybody's either charging exorbitant amounts of money to get to their APIs or they're just not allowing people to do it at all. So there's not, like, a single program that can post on all these things and it's a freaking nightmare. Like I'm not, I'm not doing it, I'm not doing it.

Speaker 3: 5:00

Never say never, but I'm not going to do this. Where's the Roku of social media anymore? But yeah, like you said, the point with the APIs.

Speaker 4: 5:07

Before every sorry, before every CEO decided they're going to charge $5 million a month to access their stupid APIs Like they were there, those, those, that software was there, run by volunteers a lot of it, but no, not anymore. So it doesn't exist, and that's my soapbox.

Speaker 3: 5:24

Yep, it could be all different next week.

Speaker 4: 5:27

We can have the same threads. Tim, are you going to do it?

Speaker 2: 5:29

No, I just, I just heard what it was this morning so no Instagram, but not yeah.

Speaker 4: 5:35

Is it Instagram or is it Twitter? Because it's run by meta.

Speaker 5: 5:38

It's run by meta, so I think it's yeah, it's similar to Twitter, but it's tied to your Instagram account.

Speaker 4: 5:44

So yeah, I've heard there's a lot of issues, and then, if you want to, get rid of your threads account.

Speaker 5: 5:48

You have to delete your Instagram account.

Speaker 4: 5:50

It's the same Very bad, bad, bad, bad, bad, bad, bad, bad, so bad, so good luck everybody.

Speaker 3: 5:57

Before we get into potential social social media CEO cage matches, because that'll be a whole, completely different episode. Yeah let's get into our topic. This evening we are going to discuss something called media networking, and we have brought on Josh Warcop to discuss this topic. Josh, can you introduce yourself and tell us what we're really discussing tonight, what it means and why it's important?

Speaker 5: 6:26

Sure, I will try to answer at least two of those three things. We're talking about media networking. My background has always followed a bit of audio stuff, so I was a DJ for a while. Really like DJ and really like doing that, you know, having fun partying etc.

Speaker 4: 6:41

We're gonna have to talk about that later.

Speaker 3: 6:43

Oh, we didn't. We're gonna talk about DJing. I was just a DJ once upon a time.

Speaker 4: 6:47

Didn't know that. No big deal.

Speaker 5: 6:49

Oh, no big deal, yeah. So the audio stuff has always been like a hobby, right, and so the challenge in that is like okay, so the hobby is great, but I need to make more money. So let me figure out this engineering and networking thing. I got into tech a long time ago and kind of like, how can I kind of blend what I like on the audio stuff but also do networking and you know some computer science, tech stuff and early 2000s, right, cisco came along and says, hey, we're gonna do this thing called voice on networks. They bought a company at the time. They're like and the bar that I was working at just doing you know basic server admin type stuff. You know they're like hey, you want to go learn. You know Cisco voice. I was like sure that sounds fun, it's audio. You know how can I blend some of that? And so that kind of kicked it off. Right, you know how can I, how can I put audio and learn how to deal with phones, deal with audio inputs onto networks and get that stuff converted into wonderful IP packets? That's been kind of the career kickstart was like.

Speaker 5: 7:46

So I kind of followed audio and then later video. As you know, bandwidths went up. You know ethernet speeds went up as like now we can do more video and video and video. So I kind of have just been following that. That's kind of a you know short story, origin story of how I got into voice and video. Of course, you know, did the Cisco collaboration thing for a long time and then now it's kind of pivoting a little bit again. Right, there is a you know a broadcast TV industry that has been around a lot longer than ethernet's been around. You know we're talking, you know 1940s, right, when NBC and CBS got started and how they did analog video etc. Then along came digital video in like 1989, 1990, when those standards got written and everything went digital. So now we've had digital video, we've had digital audio for many, many years and the next pivot that's happening is okay, cool. How do I get away from even the digital video and just use ethernet and IP networking for everything? That's the topic.

Speaker 4: 8:43

So currently, or at least I guess up until recently, ethernet was not being used to carry that data, at least broadcast TV stuff.

Speaker 5: 8:53

Right, it was not. So STI has been around a while from a digital video perspective, and so if you see a lot of like the cable picture stuff from like the Reddit threads and stuff like that of like the massive amount of cables that are like in Iraq talking thousands of cables a lot of times it's actually video distribution. Cables could be camera networks, could be video distribution. Because it's not a, it's not an ethernet switch that's doing all that stuff. It's usually a back of a video router. When you see like the 1024 cables plugged into a rack, that's that's STI, that's digital video. And so the pivot is okay, cool, sti has been great, but now the bandwidth requests are going up even more. Sti was like at 1.3 gig. I'm going to get all this math wrong. So if anybody wants to fact check me while we're talking, that's totally fine. We get all the math wrong.

Speaker 4: 9:41

We are probably intentionally. I'm going to Google it right now. Let's go.

Speaker 5: 9:45

So STI right, uncompressed video is 1.3 1.4 gigabit per second, and so you already can hear the problem of that. Trying to put that on ethernet right, that's faster than a one gig ethernet port. So video and STI videos has been around a long time and it's probably not going away anytime soon. But how can we start leveraging faster ethernet speeds to pivot over to ethernet, and that's what the broadcast industry is challenged with right now?

Speaker 5: 10:10

I was fortunate enough to make it out to NAB show this year. Nab show is probably like three times the size of Cisco Live. They fill out the entire Las Vegas Convention Center. All three are like North, south, east, west Wings of Las Vegas Convention Center, and the common theme at NAB show this year was definitely continuing down the path of well, how do I put all this stuff on a network? Right, servers with network ports, servers with GPUs running network ports and they're not shy about how fast they need to go either. They're talking 100 gig ethernet ports, right? They're like why don't we just go all the way fast so we don't have to have this conversation again?

Speaker 4: 10:49

That's the direction. What are the limits of STI? I assume we're hitting up against limits of STI and that's the reason for doing this right. So do you? I don't mean to quiz you, but I'm just curious.

Speaker 1: 11:01

You're curious what are the limits?

Speaker 2: 11:04

I'll try.

Speaker 4: 11:05

Do you know the limit of STI Like, at what point can you say, okay, this will not travel well over STI? We?

Speaker 5: 11:12

can't use it. Yeah, so STI has generationally been a thing, right? So they started HD STI, then we had I think it's 6G STI and then 12G STI and then 24. So those are measurements of the gigabit per second. So they also had some concepts of, like, how can I do dual cable STI, right? How? Basically well, I'm going to say it, wrong again. It's not multi-channel ethernet or a lag or a trunk port, right, but it's a similar concept of, let me, let me put it on two cables instead of one cable, and then they've done that a couple times.

Speaker 5: 11:41

So we're reaching like a limit, absolutely a bandwidth and just ability to to not only push the bandwidth but also all the additional metadata that I now need to send, like we're familiar with Netflix and the YouTube's trying to do HDR content or Dolby vision or you know that high extra color bandwidth Um, all of that's more data, right? So we're definitely pushing the limits when it comes to what SDI can provide for us. It's not going to weigh anytime soon, right? I mean, there's still like a broadcast standard to reach the masses, right, we're not all going to get 4k video. Everybody that's still getting their TV over antennas, right? That's going to be a thing for a very, very long time because of just the way the world works. So we still need to meet those needs and those needs can easily be met with SDI. We've been meeting those forever.

Speaker 5: 12:27

Uh, but in the newer HDR 4k, maybe even 8k down the road, right, that's going to be a lot more bandwidth required.

Speaker 3: 12:34

So you mentioned the. The big challenges is bandwidth, so it's pushing us from these localized digital uh media cabling and switching to Ethernet and IP networks. So you're, you're taking an industry to traditional data networking. That's never seen that before, or at least to the the scale that most enterprises use it at. So that's to me that's screaming skills gap. So you have a need for networking minded people, network admins, network engineers into the video space, but they kind of need to be able to speak some of that video language as well. So how, how do we get those network minded people into these media networks?

Speaker 5: 13:18

You're right, that's a. That's a big question mark right now. I think a lot of companies and TV stations and broadcast stations are trying to figure this out right now. Right, I mean the standard's relatively new. Right, I mean just because the standard gets written doesn't mean it's going to happen the first year. Right, it's going to take some time for it to get productized into stuff actually in the field. It's wrong the first time. There's a whole life cycle of the stuff. So we're really reaching a point where those skill gaps are getting challenged. So we've got a lot of broadcast engineers and AV enterprise engineers that are trying to learn more networking and we've got a lot of network engineers that are trying to learn more broadcasts. It's like it's like a lingo problem too. I mean they've that industry's got so many more acronyms than you think networking has. It's amazing.

Speaker 4: 14:06

Who do you think is going to have the hardest time? Like crossing over network engineers into broadcast or broadcast into networking.

Speaker 5: 14:13

Oh man. So I'm going to tell them myself. Right now I'm having a challenge learning the broadcast stuff. I think it's going to be easier for the broadcast community to learn network engineering. Quite honestly.

Speaker 4: 14:22

Now, why do you?

Speaker 5: 14:23

think so Easier, not necessarily correct.

Speaker 4: 14:25

Wait, can you elaborate on that a little?

Speaker 5: 14:28

bit yeah, yeah, so so much if you want to. I think it needs to be a dual role in the future. Instead of saying, hey, a broadcast engineer has to know everything. Right, how do we embed a second or a third person that knows the specific skill set, right? We're getting into this too much knowledge domain problem. It's almost like saying I know everything about cybersecurity. That's not a thing, right. It's like saying I know everything about broadcast engineering is not a thing right. It's like I know the broadcast engineering side and then we need to, or AV engineering and I need to add in a network and then also a cybersecurity person into the conversation, right, yeah? Somebody just said in chat yeah, you're right, it's a lot more cost. It is driving up the cost by adding people into the conversation, but I think a lot of the companies that are doing this have the dollars to spend on it. We're talking some pretty big media conglomerates and they get a lot of advertising revenue.

Speaker 4: 15:19

I can almost see, though, companies, more than maybe like hiring extra people, potentially actually like asking their current people to start expanding their skill sets. Is that something that lines up with your experience so far, or it does?

Speaker 5: 15:36

Yeah. So at I guess I'll give an experience from NAB show I actually put on a vendor shirt and stood in their booth and kind of, you know, talked about people coming up and talking about this stuff. It was a networking company and there are a lot of broadcast engineers coming up going, hey, where do I even start learning Ethernet and IP? That was a very, very common conversation, because they've got the background right. They know how to sit in the control room and go live and do stuff like that and do what they do. I don't know what they do. I haven't done what they've done, so I can't elaborate on that, but they're asking that for sure, because that is what's happening. You're exactly right. The companies are going hey, look, we're making this pivot over to IT infrastructure servers, router switches. You need to start learning that and I feel like that's a problem that companies are putting on people, right, I don't know how we're supposed to know everything. Yeah.

Speaker 4: 16:27

Honestly, it kind of smacks of the like you got to learn automation now.

Speaker 3: 16:31

I was just going to say that, sorry, no, no, no, no, I love that we're thinking the same thing, because I'm drawing a very similar parallel from, you know, broadcast engineering people having to learn networking to network engineers having to learn things Like automation and cloud. Yeah, it's just different problems but a lot of similarities, it seems. Oh yeah.

Speaker 5: 16:52

Go automate everything tomorrow. You can do that by next week, right? Yeah, so you're absolutely right, it's becoming a too much knowledge domain problem and I think companies are going to have to start addressing it legitimately, because you're going to start running into personality problems of the engineers that haven't learned yet to say no, that's not what I know and I don't have time to go and learn that. People want to think they know can learn everything, and that's just going to be something we're going to have to address. So you're right.

Speaker 5: 17:22

It's a skills gap problem and you know that's kind of want to want to. Part of the reason to get on here and talk about it is like how do we go after it without saying, hey, you got to know everything?

Speaker 4: 17:32

Can we talk about some of the specific technologies that somebody who's a network engineer now maybe, or getting into network engineering but is actually going to be a little bit more complicated? I'm actually interested specifically in kind of plugging into the broadcast world or like the voice and video world. What kind of protocols, technologies, cert, track, whatever, Like what should they be looking at? Do you have like a list of things or any recommendations?

Speaker 5: 18:00

Yeah, I think a good recommendation is if you're already down the path with any sort of Cisco certification. There's some books that have been around a very long time about how to put voice on networks. The broadcast industry and even the radio broadcast industry did it before video broadcast industry did. So radio was very quick to pivot to this alongside voice Right. They were like audio is easy, right, it's a 64 K stream that I can drop on the network, no problem, we do that all day long. So audio broadcasting and audio and phones on networks were definitely a thing first.

Speaker 5: 18:38

So there's a lot of books that already exist on how to do that and how that works. Right, how do I do packetization of voice into an IP packet? Right, and how that process works. There is a CCNA or I don't think it's called CCNA anymore but there is a voice level one book, basically from Cisco collaboration that is. I think it's still a good entry point from a network engineers perspective looking at this, because it's going to give you some fundamental concepts right Of like how this actually happens. And those same concepts, I think, translate to broadcast, because it's still the same RTP packet right On the network. We're just doing it a much higher bandwidth when it comes to video.

Speaker 3: 19:20

So we're talking about dropping these voice and video packets, these streams, onto IP networks. Is this something that we're combining on traditional data networks? Are we treating this kind of like OT and security camera networks, where they're separate physical networks? What does that look like?

Speaker 5: 19:39

Yeah, and it depends on the venue, depends on the building. I'll give an example of a live venue, sports venue, right, we can do, or it is possible to do, that on one converge network, when it comes to audio only. So if I'm or if I'm talking about all of the digital signs, or you go to a baseball or football game writing you see all the signs that you know, will you walk out in the hallway and like hey, there's the game on the screen, right, I can do all of the voice and video and the compressed stuff on one converged network. Where you can't do it on a converged network is when I go up into the broadcast booth and I need to get the broadcast cameras up into the control room and then out onto the scoreboards, right, that's the high bend with stuff that has to be a separate network because it just you can't. It's too much bandwidth, I can't run that over one gig network and I'm probably not going to spend the money to run 100 gig for all of the stuff that doesn't need 100 gig throughout the whole venue.

Speaker 5: 20:35

Even in an enterprise it's usually a separate network. So if I'm doing digital signage or video distribution through the building, it's a lot of times a separate network, and the reason for that is it comes down to who wants to manage the gear number one and who wants to manage all the multicast required. It's all multicast we're talking about, right, it is. It is a multicast network 100%, and a lot of times the engineers that are running the corporate network are like no, you're not going to put that stuff on my network, forget it Just go build your own other network and go do that over there.

Speaker 4: 21:10

Every network engineer that I know loves multicast. It's their favorite kind of traffic. Yeah, oh yeah, everything we're talking about.

Speaker 5: 21:20

So far is multicast 110%. Yeah, I mean, there's obviously some unicast when we're talking about just audio, but when we're talking about a video distribution, yeah, multicast, so multicast would be a good thing.

Speaker 4: 21:33

I assume, like PIM and IGMP, like at least the base six, if not like, that is, core advanced stuff would be very very good, okay. Very good for people to study if they're interested in going this route with their network engineering Okay.

Speaker 5: 21:48

Cool, yeah, absolutely. I mean multicast and PIM and how to manage that on a network is essential. I would say that's one thing to help, like when the AV integrator shows up at your building and wants to plug in a whole bunch of stuff that's your, you know AV tools or conference or and stuff or whatever like that, it's super important to know what the multicast implications are, because they're going to want to plug it into your network and you better go.

Speaker 2: 22:15

I either have a handle on this or I don't Boom broadcast.

Speaker 3: 22:18

Yeah, yeah when we talk about enterprise networks and creating these separate physical networks for media, for voice and video. If the traditional IT network teams aren't managing it, then who is? Is it typically integrators or just other? Like we talked about the split between IT and OT, we've had a recent episode about that, like who would be managing and maintaining these networks.

Speaker 5: 22:47

Well, I mean it depends on the customer, right? Sometimes the integrator will come in and do the days of your day one, get it up and running, everything looks great, and just basically handed back. It's not a network that's touched very often. I would say it doesn't have the same life cycle as an, as a corporate network for desktop servers etc. Probably has a much longer life cycle because it's very purpose built and so I'm going to put it in the closet and it's going to do its job for moving video in my building probably for the next five to seven years. So the IT team can patch it, secure it and isolate it. But I don't need to go back to it very often. So some customers, it'll just get kind of handed back to them.

Speaker 5: 23:25

In other cases, you're absolutely right, it may be the integrator that's going to do the day two, support and just kind of care and feed. Again, it doesn't need a lot of care and feeding. It doesn't need some from a security perspective. But I mean there's a variety of customers out there, right? Some folks want to take it on, some folks don't want to take it on. But it is a separate, typically a separate video distribution and as we get into the higher HDR slash 4K content that wants to get pushed throughout a building. Right, we're starting to build 10 gig networks. If I'm doing IPTV through a building or a prison, that's going to be a 10 gig network. These days and it's usually just got its own separate care and feeding.

Speaker 4: 24:07

Did you say a prison, just there?

Speaker 5: 24:09

I did say a prison. Yeah, that's actually so are we talking like?

Speaker 4: 24:11

security cameras.

Speaker 5: 24:13

Yeah, no, no we're talking about prisoners watching TV. Okay, yeah, yeah, absolutely. So if I wanted to put TVs throughout an entire campus, it is a campus, it's a campus network, it's just a prison. Yeah, it's a lot of TV Interesting. That was the next thing I wanted to talk about.

Speaker 3: 24:30

Exactly, and on the enterprise side we talked about things like, like stadiums and now prisons, but what are some of the? What are some of the other examples of enterprises that would have these things? And I think you just kind of answered this question in IPTV. Can you kind of get into IPTV what the use cases are and why we would have a separate media network for IPTV?

Speaker 5: 24:59

Yeah, absolutely. So the thought process is, before this was landing on the Ethernet right, I would run a coax distribution through my campus. It could be a college campus, it could be an education facility. That's one big concept for this right Of how do I get TV into all the dorm rooms or TV into all of the classrooms that need to display the news. Before that was networked right, that was a separate coax based, you know, RG6 network that ran through everything. So instead of having that additional cable plant, well, let me just don't do that anymore and run it on Ethernet as well. This is another case where Ethernet is winning in a vertical right, and so that's that's kind of concept. So education facilities, corporate distribution of TV, like if I want to show a news channel in a break room. Some of it isn't a lot of TVs, but IPTV has definitely been a driving force behind getting everything over to the network.

Speaker 4: 25:54

Would you say a lot of companies. So it the impression I've got is that we have sort of we're in sort of this like in between area where we've got like older technologies being used to carry a lot of this and we're trying to move towards the newer, like our you know, relatively newer Ethernet, would you say. I'm assuming the majority of, like, broadcasting companies and things like that are using older technology and slowly moving in the newer technology direction. Is that, am I? Am I?

Speaker 3: 26:24

right about a lot of like retrofitting happening.

Speaker 5: 26:28

There's definitely a lot of retrofitting. There's a lot of converting going on. We talked about SDI earlier. Sdi to IP converters are a huge thing right now. Of like, how do I get this existing stuff in a little box that converts it from SDI to IP and that's all it does. You've got one video stream in one side and an IP stream out the other side, and how to bridge the gap is a big topic of conversation right now. There's a lot of use cases and studies going on of like what's actually cheaper in the long run? You know people are looking at the financial side. Nobody really knows, because it's all getting productized right now and ship it. Ship it by this, by this, by this, so it's, it's like it's. I can't use the word exploding. I was told to stop using that because it implies violence. So there's an explosion of Ethernet happening, but yeah so, no, it's all good.

Speaker 5: 27:22

So, yeah, it is definitely a rapid expansion for some people. A lot of people are taking the pause and going is this really financial, you know, justifiable for our station, our local station group? Is there a reason to do this the way that we've? You know, we've been doing it this way for 40 years, so why do I need to switch right? What is the real advantage to this? And the advantage to it long term is probably how content is going to become interactive. I think that's, you know, a big change of like hey, we've been delivering video content to you, but how do I start bringing in all the other content delivery streams and advertisers that are now all IP based into that? So the content delivery market is definitely driving a lot of this too. Of hey, we need something newer, better, more modular you know one way to say it of how to get content and advertisements into that.

Speaker 3: 28:14

Leave it to the hyperscalers every time. Yeah.

Speaker 5: 28:17

Yeah, leave it to the hyperscalers.

Speaker 1: 28:18

And now a work from our sponsors. Hey one fans. Unimus is an on premise network automation and configuration management solution designed for fast deployment and ease of use. On top of disaster recovery and change management, unimus also handles config all day. The config search feature lets you easily look up and list all the devices in your network with a particular config. By querying command outputs from your devices at scale, unimus also gives you visibility into the running state of your network. So if you're interested in a device agnostic NCM software to streamline and automate configuration management for your entire network, check out Unimus. Now back to the show.

Speaker 4: 28:56

So are people building like still building SDI based infrastructure, or is it mostly the new stuff being built as internet, ip based?

Speaker 5: 29:06

Yes, yeah, there there's definitely a lot of customers that are still building SDI infrastructure and the reason for that is number one cost. Right, it is a bit expensive at the moment, depending on who you ask. Right, I'm going to get some. Somebody's going to come and tap on my shoulder. I said that, but depending on who you ask, yeah, the newer Ethernet infrastructure makes sense for maybe some larger customers, whereas a smaller station might need to stick with SDI because of costs. They know they don't. They don't necessarily need to change right it's. It's a. What are the? What are the core business values that I'm getting out of this instead of just changing just to change right? That's like the whole automation story we just talked about. Do I really need to automate or just need to hire a person to punch the buttons?

Speaker 4: 29:52

That's fair. Yeah, I'm assuming it would be like mainly fiber based stuff and that can definitely be expensive, so that makes sense.

Speaker 5: 30:01

Yeah, I think the, I think the Ethernet market to is helping drive. This is now becoming cheaper to go faster, right, it's. It's like you know, we used to look at one gig to 10 gig was just expensive, right, the, the, the chips were not developed to go faster without spending a whole bunch more money. Now it's like 400 gig, no problem. You know, 800 gig, no problem. Right, I can get 32 ports of 800 gig and one are you now? So why wouldn't I just go ahead and buy an 800 gig switch in one, are you? As compared and this is, I think this is a good comparison to I can get that much bandwidth and one are you? And an Ethernet switch to get that much bandwidth in a traditional video router is like half a rack, right, it's.

Speaker 2: 30:45

There's like a you know power economics problem.

Speaker 5: 30:49

There's a cabling plant infrastructure to get that much bandwidth in the same older form factor. You're talking a lot more physical space to make that happen. So a lot of economics are coming into play. And how can I go faster but not take over like another 10 racks of equipment?

Speaker 4: 31:03

Yeah, ok, you're an architect. Somebody comes up to you and says retrofit my STI based like infrastructure. How do you in general go? What are the things you have to think about there, like what are the, I guess, gotchas with STI going over?

Speaker 5: 31:20

to you know, IP based stuff.

Speaker 5: 31:21

So the, the gatchas, would be OK. What? What do you want to do with the? You know you have to evaluate, or think about the cameras that are involved. You have to think about the video routers that are involved. You have to think about the control room and their workflows. You know they're there. You know when, when, a when somebody goes live in a control room, right, it is a live production. There can be no problems. So there's even a standard written to duplicate the network for live broadcasts. Right, don't build just two of these or just one of these. Build two of these and we're going to put the video stream twice.

Speaker 5: 31:56

When I talked about earlier, there was the older STI standard that was like let's put it on two wires instead of one. Well, that still kind of permeates the solution and it can drive up costs. So a lot of times we're talking budget and financials way early in the conversation of like okay, look, this is how much it's going to cost. This is how much bandwidth is coming off the cameras. How's that going to impact all the port speeds that are needed? It's a pretty intense conversation.

Speaker 4: 32:22

Are we talking like two, potentially two completely separate active networks? Carrying the same data at the same time.

Speaker 5: 32:33

Carrying the exact same data at the same time. Exactly Interesting, Just for the potential of any failure during a live broadcast right. That second video stream is already being from the camera to the video router. I've already got a second network. It's flowing through. That's wild. It's two data center networks.

Speaker 4: 32:53

Okay, so why that? Sorry, this is very interesting to me, right? Why take that approach for redundancy as opposed to like lag or like redundancy in smaller ways, more granular ways within the network, like an active router in a standby route, that kind? Of a concept. Why two whole separate networks?

Speaker 5: 33:14

Oh, that's a great question. I don't know. I don't know the core answer of why that standard was written. I would love to reach out to the people and go why did you write that standard? This is two data center fabrics and we're talking about data center fabrics. We're talking about definitely 100 gig like spine leaf fabrics, right, typical data center spine leaf networks. They're not running VXLAN, they're not running EVPN or anything like that. It's basically a full layer three network from the port. So the port is actually a layer three, port, layer three, all the way through the fabric. And the dash seven standard is basically say, okay, build two of those and the camera's going to plug into two different leaves and the video router is going to plug into two different leaves and it's going to be the same video stream across both of them. There's just a, there's just a. No tolerance for any.

Speaker 3: 34:01

Yeah, it's just a resilience resilience right.

Speaker 3: 34:04

Like drawing and why I think, lexi, I'm with you. I think this is for like a, an enterprise networking person. Why I think this is kind of crazy is, to me, drawing a parallel would be on a on a floor, on an access. You'd have multiple access layer closets and you would have somebody's PC that has two nicks and plugged into you know diverse switch stacks essentially. But yeah, it sounds like it's just the like you said. There's just no tolerance for for failure and you need that level of resiliency in these kinds of networks.

Speaker 5: 34:37

Yeah, and I think it it it brings up a really good point. Back to the skills gap. Right? The concepts that you just just mentioned, right, you know lag or LACP, or having the ether channel conversation those are concepts that are new to the people that are doing those designs and those standards. Right, they're not in the weeds every day of what that means and what that looks like. So it's, why don't I just build a second? One is probably the first generation of this problem. And then they're probably going to back down and say, well, if I want really redundancy at the access level, I can probably do this instead of two switches.

Speaker 4: 35:11

Yeah, I'm wondering if that's something that came from the old old days, before Ethernet even.

Speaker 3: 35:16

Just cultural differences.

Speaker 4: 35:18

Yeah, yeah, it's really interesting to think about. But I can also see how for very, very, very critical networks. Right, like you're, you know you're live it it also does kind of make sense to me in a way that, like you would have to completely separate. It's an active, active, like you know, one fault tolerant network. It's really, it's really cool.

Speaker 5: 35:40

And so and I've got a great parallel to you know some of the some of the work that that we've done with some very large sports dual events. Right, we've done some work with what was it World Cup we were talking about earlier. We've done some work with Super Bowl and we're doing some work with some tennis championships and the idea is that it's a completely there's two networks to run all of that and there's probably four networks to back all of that back to between countries.

Speaker 4: 36:10

I was going to say like when it's something like the World Cup and it's like between countries and between stadiums, between countries, a lot, of, a lot of networking happening on that.

Speaker 3: 36:21

And are there? Are there challenges around? Because those are? Are live events right? So are we. Are we streaming live media and recording at the same time? What is that like?

Speaker 5: 36:34

Absolutely so. It's all 100% live, right. So the camera feed is dropping onto the network and then I've got recording solutions that are also on that Spine Leaf Network. Those recording solutions not only capture the video, but they're also the way the servers also do the re-injection of all the graphics and all the scores and everything that you see on TV is happening on the network. It's happening from servers with GPUs integrating with graphics. So, yeah, I mean it's, it's definitely a big recording server and storage solution. Behind all of that and all of that is another thing that's pivoting to Ethernet. Right, this whole thing has not been Ethernet storage previously. Right, the whole industry is making the change over to like an IT data center.

Speaker 3: 37:19

Well, let's dig into a live event like that a little more. So let's say, you've got an American broadcasting company that is covering an event like that overseas. How do they do it? Are they? Are they shipping their own data center over there? How do they accomplish that?

Speaker 5: 37:37

So there's, there is a, there is a concept of remote production. Right, we're not going to, you know, not not have everything distributed from the remote country. Right, we're going to ship the data back to, probably a remote production facility or central control room, wherever, wherever the production company exists. So one of the things that that we've done and we we showcased a little bit of this at NAB show was that our our concept of putting a media fabric or data center fabric into a shipping container so that we could actually get that to a country like Qatar or Australia or wherever these events are happening. Right, and one of the things that was pretty neat, we kind of developed a folding rack solution, because there is a, there is a height limit to put racks on or things on an airplane. So if we needed to fly it somewhere, we actually needed to fold the rack in half and then encapsulate it in a box so that we could push it onto an airplane and send it somewhere.

Speaker 4: 38:33

Is this a rack that has like networking equipment already installed and it's just like we're just boop boop sticking it? On an airplane.

Speaker 5: 38:41

Yep, absolutely so, it's, so, it's, it's two data centers, right, it's redundant, redundant fabrics, redundant everything yeah, but it folds in half. Put it on a container, we can push it on an airplane, we can push it on a shipping container or ship it anywhere around the world Data center.

Speaker 1: 38:56

Yes, I saw it in chat.

Speaker 5: 38:57

It's very robust. That's a fun, it's a fun.

Speaker 4: 39:02

So you're okay, you've got your data center that you've shipped overseas right Sitting in Qatar, wherever. How does the data actually get back to where it needs to go, get broadcasted everywhere? Are you using, like satellites and RF? Are you you know? Is it going over like an oceanic backbone? Like, how does all that work? Does it just depend on the company, or A little bit of the company right.

Speaker 5: 39:23

Depending on who's doing the event. I will say that most of that is oceanic backbone right. We're doing this at probably you know, 100 gig speeds. On private backbones there may be four to 800 gig worth of traffic coming from an event like the World Cup. All of the camera feeds, everything is getting getting brought back into a central location for then distribution in the country right, or it may be, or it may be sent to other countries for their own distribution. It may cross countries to go back to another country. It all depends on the company right, and how they want to take the feeds from something like that and then send it out over their, you know, antennas. That's a way over simplification of it, a lot of networking.

Speaker 3: 40:09

I mean, yeah, that's always what it has to be, unless we want to do a deep dive into every little deep yeah, let's kind of go back to we were talking about that skills gap in traditional like broadcast engineers that are wanting to learn networking. Let's go the other way. Let's talk about network engineers that maybe need to learn more about these media applications. And I'm kind of selfishly asking this a little bit as somebody who's trying to understand a little more about collaboration. Do network engineers need to learn about the? You know the intricacies of things like codecs and encoding and decoding and the main reason I asked this is I recently took an intro to like collaboration course and I was all about it until it got into codecs and my brain freaking melted and I'm like so how prevalent are things like codecs and encoding and decoding for network engineers to be able to learn some of these media concepts?

Speaker 5: 41:16

How much of a nerd do you want to be Right? Okay, come on, we're on a podcast called. The Art of.

Speaker 2: 41:22

Network.

Speaker 4: 41:23

Engineering, that is your answer.

Speaker 5: 41:27

I think learning about the encoding and decoding side of it, like how the video is actually turning into an IP packet, is important. Like, if you want to deep dive into any sort of troubleshooting perspective, then you're probably going to need to know some of that. And specifically we're talking about when I'm thinking codecs. Let me, let's clarify what we're talking about here, right, I'm thinking of things like AV1. I'm thinking of Opus, I'm thinking of HVAC or Impeg, so those are, or JXS, which is JPEG XS, right? So those are encoding strategies for taking the video and audio, compressing them, if I need to at some level, and then turning them into an IP packet. I think understanding the codec process and what the codec has to do to do all of that work will make you super dangerous. In the conversation Somebody mentioned in chat it's yeah.

Speaker 3: 42:20

I was afraid you were going to say that.

Speaker 5: 42:22

Yeah, no, it's not. So it's definitely not learning the software development side. I'm not saying go to the GitHub repo and learn how to read the C and understand that, but understanding the concept of how that codec and what that codec needs to operate is probably the critical piece. Do I need a GPU in a server or do I? Can it be done on an Intel processor or can it be done with a DSP in a router? Those are where the codecs have the capability to operate. Hardware goes faster, right, I can do a lot of things in software, but when it comes to video compression and decompression I probably need a piece of hardware to do that.

Speaker 5: 43:00

One of the big things that I used to run a series called codec wars and I think I had that in the notes earlier. There's been, there's a couple prevalent codecs that are there to kind of taking the lead and it's really following the open source tradition I think. I think we all know and see that open source kind of wins a lot of times in the end. So anything open source I don't have to pay licensing Probably a good thing. Av one is probably heading that direction of being more of the winning codecs out there. If I can compress and decompress and I don't have to pay a license fee to 20 companies that own the old stuff. That's great. Instagram recently converted over to AV one for their reels, right. So anytime you watch Instagram reel, that's an AV one decode on your phone. And that happened because folks like Apple and Intel put AV1 decoding in their processors, so AV1 is a huge conversation right now.

Speaker 4: 43:54

Just real quick for my own edification. I'm trying to. So I'm deep into Ethernet right now. I'm trying to sort of relate this to the encoding that happens with Ethernet frames, right, because depending on your flavor of Ethernet, you have a different kind of encoding. So I'm trying to figure out like what, at what level of the OSI model maybe, does this encoding happen for, like video, audio compression?

Speaker 5: 44:19

I forgot my OSI model, so forgive me on that one.

Speaker 4: 44:23

Above transport.

Speaker 5: 44:25

So yeah, the encoding.

Speaker 4: 44:26

Okay, so it's higher than layer four. Okay, so I was wondering if it was like, maybe is it, is this a presentation layer or whatever the useless ones above. You know, is this one of those layers, as opposed to the lower layer, like layer two, encoding that Ethernet.

Speaker 5: 44:41

Okay, it's higher up, so so session establishment would relate to things like SIP and SIP TLS. Session establishment would also be like a HLS or how I'm doing RTMP out to YouTube or this platform that we're on right now. Right, that's more of like a streaming session enablement, the video encoding, audio, coping happens before that. Right, I've got to, I've got to turn it into a piece of data before I wrap it into some sort of session creation and then it goes down stack.

Speaker 4: 45:09

Okay, so you've got like your application recording this stuff. Then it's compressed at some point after that in one of those crazy layers, and then you've got the transport layer, and then you've got it's, you know, further encapsulated until you get down to Ethernet, and then finally it's sent over as bits. Is that like a very quick like, okay?

Speaker 5: 45:25

absolutely fairly accurate. Cool, all right my brain feels better now.

Speaker 4: 45:28

Thank you.

Speaker 5: 45:30

Now. Now if you, if you, really want to take another deep dive on this, there is a layer two video standard called AVB and there is an AVB working group 802 dot working group for AVB and their goal is to do a little bit lower OSI layer of what we're talking about. Right, how do I do media transport? How do I do timing? How do I do all of those things? Maybe closer to that, transport and layer two aspect of this, avb has been around a long time. There's actually a standard written on it and I don't. That's a challenging conversation because there's a lot of products out there that rely on AVB right now. I don't think we nest. This is personal opinion. I don't think we necessarily need AVB at that level anymore because there's just so much more robustness now, higher up. I don't need to do it at layer two anymore. I just don't need to.

Speaker 5: 46:25

I need it above layer three so I can route the thing and so I can multicast the thing. I don't need it. Talking ethernet frames Personal opinion Sorry for anyone that's in the 802 working group on that. Yeah, let's let the switches alone. Yeah, I totally agree with that chat message. Let the switches be switches. We don't necessarily need to do this other stuff too right.

Speaker 3: 46:50

So, along along the skills vein, let's say I'm a, I'm an enterprise network. I've got a completely separate physical network for my media. It's all on its own 100 gig backbone, all that good stuff. That means I don't have to care about QoS anymore, right? And if I do, how do I care about it? When it's all, it's all media. It all needs priority, it all needs preference.

Speaker 5: 47:17

Tell us about QoS and media networks. Josh, You're bringing up the QoS thing. I was hoping we were going to dodge the.

Speaker 4: 47:23

QoS thing, but hey, that's okay.

Speaker 2: 47:25

We're here now Without it.

Speaker 5: 47:26

Yeah, my, my, so. So one of the things obviously, being media networking guy over the years, has been a lot of QoS conversations. This comes up all the time. Do we need it, do we not need it? I would say in a lot of circumstances 10 years ago, absolutely, 100% right. I'm trying to put everything on the network. Everything is one gig. I need a way to prioritize even a 10 voice streams over an ethernet port. Now that it's at 10 gig and 100 gig and above, it's becoming less of a priority. But you're going to get my personal opinion. My personal opinion is, yes, you do need it in all cases, but you need to do it right. Bad QoS is worse than no QoS right. You've got a lot of control traffic that's happening on your network. That probably needs some level of prioritization and it might not hurt to put that in a priority queue.

Speaker 4: 48:20

Can we quote you on that Bad QoS?

Speaker 5: 48:22

is better than no.

Speaker 4: 48:23

QoS.

Speaker 5: 48:24

Yes.

Speaker 4: 48:25

I would like to tweet that I'm going to use it, yeah, yeah.

Speaker 2: 48:29

No that's interesting I even said it wrong.

Speaker 5: 48:31

but hey, the idea is right. You got to think about the traffic types on the network for sure, and if you've got control traffic, you absolutely have control traffic on your network. You know how your switches talk to one another is over a control plane. Are you prioritizing that and making sure that any bandwidth over subscription is not crushing your control plane? I'm going to get so beat up on this. By the way, please feel free to tweet me at atwaracop and say, oh my gosh, he's saying you probably need QoS.

Speaker 4: 49:06

Now, when you've got a network like that and you're hosting this huge event like, for example, the Super Bowl, and it happens once a year or whatever it's not happening all the time Do you make specific changes to the network for that specific event and then roll them back later? Or how do you? Do you just expect that is the network designed so that at any point, technically, it could host a Super Bowl and be fine? Or like what are the preparations for something like that besides, like shipping your data center in places?

Speaker 5: 49:35

Yeah, so the in-venue network, so like what's in a stadium or what's in a sports venue, something like that, right, that doesn't change very often, right? How the video gets to all the screens, how the phones work, how the scoreboard works, that's all fixed. That's not going to change. It's set up for multiple events, just just. It's not the Super Bowl, but hey, I'm going to have a football game here in every few weeks. Right, I'm going to have a concert here, I'm going to have all of these different things, right, the venue has to make you know they're in business to have events, right? So they've got to keep doing stuff. So the in-venue stuff doesn't change that much.

Speaker 5: 50:11

What happens for a live broadcast is all the trucks roll in the back, and so now I've got to hook the internal network out to all of the broadcast takers that are in the back parking lot. That used to be like running a whole. Well, it still is. Let me. Let me clarify. It's still a lot of cables, right. Cables going out the back door or going out into the broadcast slot, cables, cables everywhere, right? If you want these different feeds, I've got to give you five cables to do this.

Speaker 5: 50:36

How the conversation is going is now. I can run a couple of fiber connections out to you that may be 25 gig, 100 gig, and you can kind of pick it like a channel on your TV, like what feed do I want? I don't have to give it to you all on separate cables. I can give you an ethernet connection and you can view all the multicast streams in your truck and then you can do whatever you want to from there Send it out on satellite, send it on another network, etc. Etc. So the in-venue to the external venue is definitely becoming more ethernet too. Less cables, less cables to worry about and in some cases even more redundancy, because I can do it all on two cables for all the feeds that are coming out of the venue.

Speaker 4: 51:18

That's another advantage of moving to ethernet, huh.

Speaker 5: 51:20

Yeah. So there's some flexibility there and there's also just the reduction in time to set up and tear down. Right, if I had a truck roll in and they needed everything from the venue, that might have been 20, that might have been 40 cables that someone has to patch in, and you're talking about all the network takers out back. So you're talking hundreds and hundreds of cables and hours and hours of just patching things down. So it's better now if we can go to ethernet and just say here's two fiber connections, you pick whatever you want off of them.

Speaker 4: 51:54

That's a dumb question probably, but like when you say takers right out back, who do you mean specifically, like news trucks or who's doing this?

Speaker 5: 52:05

Yeah. So news trucks or broadcast channels let's say there might be a BBC truck, right, an American sports event, right, because they're going to get some rights to play stuff on their network in the EU or any of the regional takers that are going to be, you know, rebroadcasting into their network and making sure it goes out on the antenna in the regional markets, right All of those are kind of like takers. I would say. It may be like a you know we talked about earlier, maybe like an ABC or NBC or CBS that's doing a lot of this, might be a Fox that's doing a lot of this. Those three letter media companies are pretty big at all this stuff, yeah, and they've got their own regional networks that need to receive all the content. So how all of that gets meshed around, I really don't know that much about it. I just know that there's a cable that goes from a camera up to a truck and then it goes into the ether.

Speaker 4: 52:57

It goes in the air, okay.

Speaker 5: 52:58

Yeah, interesting.

Speaker 4: 53:01

We're so kind.

Speaker 3: 53:03

We're so kind with our words. In IT, right, Everybody's either a user or a taker. I love the verbiage we use. Oh yeah.

Speaker 4: 53:12

That's a good point. I never thought about that.

Speaker 3: 53:16

I didn't want to let this episode finish without talking about this. Josh. Should we talk about digital signage in different streams that go into that? So there is this magical, mythical sphere looking thing that's getting built out in the desert in Vegas. There is, what is this thing? How is it pushing these pretty pictures out?

Speaker 5: 53:43

around the sphere. That's great. So the sphere lit up this week. It was like a July 4 thing. They turned on the Hello World and now it's this orb of delightfulness in Vegas.

Speaker 3: 53:59

They're going to add the rights to that phrase too.

Speaker 5: 54:05

Yeah, so that is a very large digital sign, very, very large digital sign. It has some relation to the big signs that go all the way around the arena. It has some relation to that.

Speaker 5: 54:23

What drives that those signs are big scoreboards. It's definitely a data center network, high-bend with multicast data center. There's just so many pixels involved and so many video streams involved that I'm talking about I'm talking about several thousand multicast streams. To power something like that it's not one multicast stream I've got to send synchronized, timed, synchronized video to all of these pixels that are on the external of a sign. It's a lot of data, right? A lot of color data, a lot of things.

Speaker 4: 54:55

Timed. The timed aspect of that interests me. How do you achieve precision timing like that? I'm super curious.

Speaker 5: 55:04

Yeah, time is hard. We can have a whole episode of time we can go timey-wimey space and time, quantum physics.

Speaker 4: 55:11

Working with rockets? Yeah, yep.

Speaker 5: 55:16

So the time is definitely precision time protocol, ptp. Ptp is a huge thing. All of this definitely has to be done with timing involved as well, and that's a lot of times. Why it's a separate network is people don't want to put PTP on their corporate network. They're like no, I'm not doing that, you're crazy. So PTP v2 is definitely a thing and the synchronization of that network is, I would say, critical. And how to get all of those packets to those pixels at a perfect time is 100% a thing, and that's definitely PTP on a data center fabric. It's a leaf spine data center fabric and PTP drives it all.

Speaker 4: 55:54

That's awesome, so interesting. I don't know enough about PTP.

Speaker 5: 55:57

I have a lot of details about the sphere right now. I would like to get some more details of that in the future. I don't know exactly how many leaves or how many multi-class flows it is, but it's very large.

Speaker 4: 56:08

Are you connected to people? If we wanted to ask someone specific things about the sphere, do you know anyone?

Speaker 5: 56:16

I do, but they can't talk about it, yet I was going to say how much can really be divulged.

Speaker 4: 56:21

Yeah, I'd love to do Q&A.

Speaker 5: 56:24

They do. There will definitely be some marketing around it, for sure, once the time is right Probably once the venue goes live and U2 is playing and they're like, hey, it's U2's first concert, whatever, I'm sure they'll do some marketing around it around that time.

Speaker 2: 56:37

But it's a data center fabric.

Speaker 5: 56:38

So take your pick.

Speaker 3: 56:40

Yeah, let's start to kind of round this out, josh. You wanted to have this episode to really show people that there's a whole nother world out there to networking these media networks. What would you say to let's talk about the two different scenarios. What is some advice you would give to the broadcast engineer that needs to start learning networking or wants to start learning networking? And then what would you say to that network engineer that sees an opening and wants to be able to provide those networking skill sets into the traditional broadcasting world?

Speaker 5: 57:18

Yeah, I'll start with the broadcast and the AV side. I think starting at the fundamental levels with a non-vendor specific certification is a great place to start Learning Ethernet learning, network Plus maybe a good path Even the CCNA or JNCIA I think I said that right or even some of the early Arista certifications would be a good starting place to kind of understand, not only from Cisco's perspective how to put voice on a network and do some collaboration concepts, but also bleeding into data center infrastructure, not all of the data center products. This is a very subset of a niche world, right. So we're not talking about learning all of Arista data center networking or all of Cisco data center networking or all of Juniper data center network. That's not the thing. It's just how do I do high-ben with multicast on data center switches? That's the thing. So yeah, I mean, starting from like a Cisco CCNA level is a huge leap forward, I think. I say that because I think CCNA is a little bit less vendor focused than I think they get credit for sometimes. So I think that's a good starting place. I mean they've been doing it for years, right? What 30 plus years of trying to teach this stuff? So that's a good starting point.

Speaker 5: 58:35

I actually positioned that a lot at NAB show a few months ago. I was like, hey, just just go after some CCNA networking stuff. That's a great starting point to learn. You know core concepts and I definitely mentioned that a lot Good advice. So network engineers learning AV stuff, that's a bit of a challenge if you're not in a business or integrator that's actively doing that kind of doing that on your own is a bit of a challenge for sure. Right, I don't. If you don't have the gear and you don't have access to the gear, it's kind of like that problem of I don't have the stuff, right, how do I learn?

Speaker 4: 59:10

What about would you recommend they become DJs?

Speaker 5: 59:13

Yeah, be a DJ. Is that helpful? Yeah, learning, learning, learning the anxiety of being live. How about that? That is you make them, make some network engineers do more live podcast. There you go, and help them understand that.

Speaker 5: 59:31

Look, this thing's live. You can't have a problem. There's that's not even a remote possibility. So understanding of how to build in redundancies, I would say, is a huge part of this too. Like and not be, not be crazy about redundancies, but some level of understanding. What do I do if something fails?

Speaker 5: 59:49

There's a lot happening in the communities. I would say a great place to go for network engineers that want to dive into like a community side of this would be like the video engineering community. That was, I think there's still propped up off Reddit. You know that's a huge conversation topic right now. But if you go look for the video engineering community and if you go look for the AV integrator community, that is a great place to hang out and kind of start learning the lingo of what in the world all of these folks are talking about.

Speaker 5: 1:00:20

That's what I've been doing. My background is not broadcast at all, right, so I've been kind of hanging out in the community and jumping on you know the channels and just jump in and go, hey, what are you doing, what are we talking about today? And just trying to like learn some new stuff. I think that's a. I think that's a good way to go about it, just to get some of the core concepts down. And I was looking over my discord as I said that make sure I got all that right. I was like other community you mentioned.

Speaker 4: 1:00:46

You mentioned the NAB show, right? I'm not sure we actually said what that is. Can you just briefly explain what that is?

Speaker 5: 1:00:54

Yeah, so in a B show is basically I think I'm even going to mess this up North American broadcasters or National Association of National Association of Broadcasters, that I think that's what it is. I'm going to get beat up.

Speaker 4: 1:01:07

Okay, so it's America, it's US specific.

Speaker 2: 1:01:12

National.

Speaker 5: 1:01:14

It doesn't even say on their website immediately National Association of Broadcasters.

Speaker 3: 1:01:17

Thank you.

Speaker 5: 1:01:18

Thank you. All right, I'm going to get so hit over the head, so is there?

Speaker 4: 1:01:21

con. It's like their Cisco live, but for the NAB. Exactly so would you recommend that as like an event for a network engineer who's really interested in this stuff, or do you need to be plugged in already to that community?

Speaker 5: 1:01:34

I think it's a. So I went the first time this year. I think it was very eye opening to see like a different, different take in, different concepts and kind of what's happening in the industry. If you want to dive into, like what's happening with the standards, I'm going to give you some acronyms. Here we go SMPTE that's a big standards body that's driving a lot of this. Vsf is another one, and JT-NM.

Speaker 5: 1:02:02

Those are three different worlds of things you can go learn and read about from a from a broadcast engineering perspective. They've got some great content. Some of that content is gated, but going to NAB and sitting and seeing it like firsthand I think would is awesome. It was well worth the dollars that I spent to go out there and say, hey, I really want to learn this stuff. Let me go sit in these sessions that I have no idea what you're talking about and just start learning from scratch. That's what I've been doing Just learn from scratch. I kind of got a little bit of a background though, so I guess I'm not really learning from scratch because a lot of the concepts kind of bleed over. Yeah, I think something like NAB and going to getting some more video. Specific training is definitely available to you.

Speaker 3: 1:02:51

Awesome. So, josh, what maybe did we not cover that you think we should have hit in this one?

Speaker 5: 1:02:58

Let's take a look here. We did talk more about the broadcast side. I think that bleeds a lot more over into the enterprise AV integrator world. So, like you know conference room stuff and talking about, like Microsoft Teams on networks, or you know WebEx on networks or Zoom on networks, you know all of that is also video on networks, right, and that even comes brings us back to the QS conversation. Like, do any QS on the desktop? All of those core concepts are translated right, just different verticals. If I'm an enterprise person, I'm probably doing video through desktop collaboration or through a piece of conference room equipment. We didn't touch on a lot of that. A lot of that is becoming a lot more robust and a lot more commoditized. So for me personally it's a little bit less interesting now because it's just like get out of the box, plug it up, connect to the cloud and you have video right.

Speaker 4: 1:03:50

There's no, actually sorry to speaking speaking on. It just occurred to me speaking on like the topic of getting introduced to like the big broadcasting type stuff. Would you consider like working at maybe an enterprise but that has a lot of this video and some audio stuff going on, a decent introduction into that world to sort of slide in a bit? Absolutely.

Speaker 5: 1:04:13

Yeah, if you're in an enterprise, you likely have an AV person or lead, right. They may not be reporting into IT, but they may not be even connected to IT in a lot of ways, but you probably have an AV person that manages your conference rooms. That is a great place to go talk. Go talk about what's happening over in. What are you doing for digital signage? How are you doing video distribution in our building? How are you doing conferencing collaboration in all of our conference rooms? And then you probably have some conference rooms and maybe even some board rooms, depending on your company.

Speaker 5: 1:04:50

That has an AV rack. Go open the door and look at it. There is a whole lot of stuff in there that is not on your corporate network and that is a great place to start and just go hey, I see all these ethernet cables, what does that do? And then kind of learn from there. Right, there are some encoders and decoders in there for video distribution and that's a great definitely doorstep or introduction to it. You've got an AV rack in your building somewhere. Go take a look at it.

Speaker 4: 1:05:17

Oh, okay, so make friends with your AV person.

Speaker 5: 1:05:20

Make friends. Yeah, we definitely did not talk about how to not hate your AV people. Well, actually we did, it was learn multicast. If you learn multicast, you will make friends with your AV people, absolutely.

Speaker 3: 1:05:37

So, josh, where can people continue the conversation with you? Where can you be found on the internet?

Speaker 5: 1:05:44

So I can be found mostly on Twitter at WarCop or on Blue Sky at WarCopcom. That's the two that I'm kind of leading with right now. You can definitely find me in the communities that I've mentioned earlier, also in all of the network and in art of network engineering community. You can hit me up there. I'm mostly leaning towards Twitter and like Discord communities right now, so feel free to ping me. I'm usually at WarCop somewhere in those. I don't know that I'm not that anywhere else.

Speaker 3: 1:06:15

Lexi, any last minute things before we close?

Speaker 4: 1:06:18

No, this just like. It was super fascinating for me. So thank you so much, josh. Like I've actually learned a lot and it's making me a little nervous about not knowing much about multicast, but I'll get over that somehow yeah this is super fascinating.

Speaker 5: 1:06:34

Yeah, it's fun. It's been fun for me to learn it too. It's kind of a pivot for me and it's like hey, this is actually happening and there's some cool stuff going on.

Speaker 3: 1:06:42

Well, let this be a lesson to all of our listeners too. This episode happened because Josh reached out and thought that this would be a great thing to discuss and bring out, so traditional network engineers know that this happens. I love what you said at the end there, in that I know each and every one of you in your enterprise networks has an AV rack somewhere. You just got to go find it. So if any of you have any show ideas that you would like to come on and have a show about, definitely let us know. You can find us on Twitter and multiple other social media platforms, typically at Art of NetInge on our website, artofnetworkengineeringcom. Also, check out Cables to Clouds, our other podcast, all about going. You know the hybrid cloud environment.

Speaker 4: 1:07:33

Check that out as well. By the way, we're all on everybody's on TikTok now, so A1 and Cables to. Clouds have. Tiktoks Check us out.

Speaker 5: 1:07:44

Tim's like oh no, not that.

Speaker 2: 1:07:45

Yep, yep.

Speaker 3: 1:07:48

I don't do TikTok, but I heard that I was the face of like our first TikTok. Oh yeah, we have a great one of Tim doing an intro.

Speaker 4: 1:07:56

It's really good. It's got musical background and everything.

Speaker 3: 1:08:01

So go check it out. So we are the TikTok now as well, you can find us anywhere. Thank you all for joining us. This has been another episode of the Art of Network Engineering. We'll see you later.

Speaker 2: 1:08:38

That's Art of Net. You can also find us on that weaving web that is the internet, at ArtOfNetworkEngineeringcom. There you'll find our show notes and some blog articles from the hosts, guests and other friends who just like getting their thoughts down on that virtual paper. Until next time, friends, thanks for listening.