Chris: 0:00

Yeah, dude, not much to add here, but this is the kind of shit that just pisses me off. I feel like we're constantly talking about the same numbers in a positive light and in a negative light. And it's like we had one article just talking about their investments in AI and it's like, well, their revenues increased 90 percent year over year in Q4. And then the next article I'm like, oh, that sounds pretty fucking good. Like 19 percent, like that's. You know, as a growth company, that's pretty damn good. And then the next one, I was like, oh yeah, that was pretty bad, like we didn't live up to expectations.

Tim: 0:29

Had a terrible quarter. It's like fuck man dude Like Jesus Christ. Hello and welcome to another episode of the Cables to Clouds podcast fortnightly news. As always, I'm your host, Tim, and with me, of course, is the other guy, Chris Miles. I wouldn't go that far. We got a book out, so like he's famous now. It's true.

Chris: 1:07

He's a famous author now.

Tim: 1:09

And so he is a. Yeah, so him and Steve are famous authors and I'm just kind of tagging along with the rest of them. All right, let's just jump right into it here and not stand on ceremony. So interesting couple of weeks. We're not going to cover everything and there's plenty to cover, but we're not going to cover everything.

Tim: 1:29

So we got an article from SDX Central. It's a little bit of an opinion piece kind of, but basically the Department of Justice sued to stop the merger from HP and Juniper Merger. I guess HP bought Juniper. It's not really a merger, is it? We reported this last year. You know $14 billion, if you remember. We said how ridiculously cheap that was, but anyway.

Tim: 1:54

So it was set to close at the end of the year and the Department of Justice actually sued to basically block the acquisition and the reason stated was there's only three wireless providers like enterprise wireless providers on the market, which is Cisco, juniper and HPE. So you know Aruba, mist and you know Aeronet or Meraki, depending, and the DOJ says basically allowing the acquisition of Juniper by HPE would take it down to two vendors and stifle innovation or competition rather, or, yeah, stifle competition. So this is kind of an interesting move by the DOJ. I think it's interesting because they don't necessarily understand the wireless market, which is not really a surprise understand the wireless market, which is not really a surprise. Basically, the article is basically positing like well, will Cisco benefit from this block, from this block, this suing to block? And HPE and Juniper have basically been using the arguments against this block, saying that you know, actually, you know, allowing them to merge actually provides more competition with Cisco because they can merge their tech stacks, they can get missed in Aruba and whatnot, and they can actually be a good competition to Cisco in this market being acquired by hpe. It's actually having the opposite effect where the uh, the w land market is being split and and cisco's kind of winning by default.

Tim: 3:30

So, and it's kind of what the what this article suggests as well. So I, I don't know. I I have my own opinions. What do you think, chris? Do you think that's accurate?

Chris: 3:38

um, not necessarily. I mean, it's like I think the I think the overall kind of thought here or the findings here, is saying that you know if, if this merger goes through, then that means two companies will own 70 percent of the market cap for it, for the WLAN market, right, and it's like I don't know man, it's like that's, I don't know what that does for innovation necessarily. I mean, I think there are people that are probably in that bucket now that are either HPE customers or probably not Most likely they're Juniper customers that might end up going to something else just because of this and they probably chose not the top two vendors for a specific reason. But I don't know, I don't know if that really kind of does anything to the competition or the kind of free market, so to say, because I mean, even within those multiple umbrellas, there are still several choices able to be made there, right, controllers and things like that. There's Meraki, which is an entirely different solution.

Chris: 4:45

I'll be honest, I haven't been that close to it so I don't know how much integration they've done with each other over the past few years, but I imagine it's probably not a ton. And then the same on the other side there's. You know, hpe bought I mean, they bought Aruba years ago, right, so that's, and I definitely I see a lot of that implemented today. But then Juniper, like the whole, we talked about this back in what was it? November or whatever. Like the reason why this is happening is because, I mean, from my perspective, it's because HPE wants Myst right, they want Myst AI and they want that capability and I don't know if that's incorporated into Aruba. That doesn't seem likely to me. It seems like it will be a very like. It will be not a frictionless install to get that.

Chris: 5:35

So I don't know, like there's. I can see the justification that there's, you know kind of two big umbrellas over that 70%, but I think that's going to sway people too. Like that might cause people to like, hey, maybe I should start looking at you know, ruckus, or you know some other vendors that are in this space, right, cause I'm sure the other vendors that are like, oh, this, uh, that are in the in the wireless space, they're like hey fuck you, we're here too.

Tim: 6:07

You know they're probably pretty pissed. But yeah, I don't know. Yeah, I, I, I, yeah, generally agree with that. I mean, 70 of the market is obviously the lion's share, uh, between you know, those three vendors. Um, I'm also not 100 convinced that allowing hpe to acquire juniper does functionally, you know, consolidate the market. Like you said, there's. There's so many, even within the vendors that are out there. There's so many different types of of wireless type solutions that it doesn't even feel like the same. I think.

Tim: 6:30

You know, obviously Mist AI was a huge thing that HPE wanted. But even then, like you know, aruba hasn't really been enterprise grade, like it's really always been kind of small medium. So if HPE wanted to capture more of the enterprise market, you know the Juniper acquiring Juniper for Mists had its own value there. I don't know about them going after it and we talked about this forever ago, right, I'm not sure how much of the data center market or firewall market they're interested in. That might have just been collateral damage, like hey, we get that too. But the big one is I'm not sure that allowing this, or rather blocking this acquisition, fundamentally protects the market from lack of competition, you know.

Chris: 7:13

Right right.

Tim: 7:14

And Ubiquiti and, like you said, ruckus and some of the other ones that are out there. They're 30% of the market and yeah, I don't know, at the end of the day, I think the DOJ is thinking this is like another Microsoft type of move, and I don't think it's that. I don't think it's anywhere near that cut and dry, you know, than here. So the question that the article poses, though, basically is is this a good thing for Cisco? And I would argue that it, kind of by default, it has to be good for Cisco, like you know. I mean, like you know, because any further consolidation or cross-pollination or whatnot of a rival's tech stack has to be bad. Because you know Catalyst and Meraki, you know they've been, like they've changed labels but like the actual integration is still extremely poor to none almost. So, yeah, I mean, they still sell them as completely different sets of solutions.

Chris: 8:11

Yeah, I don't see how Cisco could come out any worse in this situation.

Tim: 8:15

Right Because of this right?

Chris: 8:17

I mean, at the end of the day, then it's still three separate products and then they're competing with Juniper and HPE, just like they were today, right?

Tim: 8:33

So I mean, yeah, it's kind of a obvious default, I think Yep, absolutely. So I am curious to see if this kind of pushback by HP and Juniper go anywhere. Or actually, I'll be honest, I don't even know what the appeals process I mean. It says sued, so obviously it has to go through the courts, right, ultimately, it has to go through the courts and, right Ultimately, it has to go through the courts. And so they've got to make their cases on why it is or is not a stifling of innovation or antitrust or whatever. So we'll have to watch this one and see, because this is not unusual for big tech companies to buy each other. So this is going to happen again.

Chris: 9:00

So long story short. This has at least been delayed by somewhere between like four or six months, at the very least.

Tim: 9:06

I bet, I bet longer than that. Yeah, I mean given the situation we're in and the big money and the big you know, the DOJ is involved. Actually, I'm curious now I have to say I don't know when this was filed. Was this filed before or after?

Chris: 9:22

New administration. Yeah, yeah.

Tim: 9:24

That might be worth. You know, that might. That's a good point, right? So the Biden's DOJ might have filed this and Trump's DOJ might be like, no, we don't care, but that big company is by big companies. You know, this is great for make America great again. So yeah, I was curious. So let's keep an eye on this one. Yeah, we'll see.

Chris: 9:44

All right. Next up two relatively quick ones for CIO Dive. We won't go too much into the article itself, but basically we have an article, each about AWS, as well as Google, about some investments into AI capacity. So AWS is looking to invest about $100 billion for increased capacity, specifically targeted around AI workloads, and kind of the same thing for Google investing $75 billion. And they're more or less following suit from what we can tell here. Because I think, if you recall back, I think on the 14th of Jan somewhere around there, we reported that Microsoft was putting $80 billion into cloud data centers for the purpose of AI. So number one in the market was doing $100 billion, number two is doing $80 billion and then third is doing $75 billion. So that's kind of what we see in terms of investment here.

Chris: 10:43

I'm curious. This just kind of sparked something in my mind and I'm sure many other people are thinking this like with the whole rumbling we saw around the deep seek models and the in the cost optimization. There it's. It's really I'm curious, like cause. I feel like that what that is bringing to the light is the, the ability to do more with less, um, or maybe I should say like the ability that you'd only have to do so, like the extensive piece to build a foundation. Llm can happen far less often than we thought it was going to happen. Right, and maybe using those foundation models to, you know, build these smaller, you know more optimized models, that that still, you know, kind of reach the same outcome, or a very similar outcome. Right, um it like, is it good enough? Uh, kind of thing. So I'm like, oh man, I don't know how much is this is actually going to come to fruition in the longterm. Um, based on how that is disrupted, everything Right.

Chris: 11:40

I mean that's, that's my kind of what's rolling around in my head. I don't know about you, tim. How do you feel?

Tim: 11:45

Yeah, no, that's so I mean these are. I don't think people appreciate how colossally large these numbers are, right, Like 80 billion, 75 billion, a hundred billion billion of the B, and we're just you're so used to seeing this number now that it's like oh, sounds like about how much you should spend. It sounds like how much you should spend for AI, but like I mean that's an absolute shit ton of money Like and where's the bomb? I want to know what does the bill of materials look like for $100?

Chris: 12:14

billion. It's all Cisco branded SFPs.

Tim: 12:18

Oh, that explains it. Oh, they're using single mode fiber. Oh, yeah, okay, that'll do it. Oh, man, no's, it's insane, right, like so, now, this had to happen to some degree. Uh, they've been. They've been having capacity issues forever, right, and just as before, even ai even came into the data center. Just, they were having capacity issues. Now, this is specialized equipment, of course, so it's not quite the same. It's not like they're competing with their ec2 instances or somebody's EC2 instances for capacity, but think of the capacity of the data centers themselves as well.

Tim: 12:53

Right, we were talking with Peter Jones about how ridiculously like, how many racks and racks and racks of power cooling and gear it takes to build one of these things, right, I do agree with you that what we're actually going to see is a splintering away from large language models into like kind of agent focused, like specialized models, and that's probably what it should have been. But you know how it is, like all technology, you start off with the generalized thing and then you, okay, now I know how that works, now we can, you know, specialize it. So, and I I'm hopeful as well as you that, uh, that actually really does end up being, you know, less tokens, less power, water cooling to. To get to the same, to get to good answers, it should right, instead of lighting up the entire model every time somebody asks a question uh, anytime someone asks a question. Rather, um, you know, you've got this very specialized model where you can just light up the parts that matter, and just physics suggests that this should be an easier thing for everybody to do.

Tim: 13:50

So again, yeah, this reminds me of I don't know why, but all of a sudden I got reminded of this old sci-fi book I read forever ago, called the Forever War. These two I don't know if you ever read it or not, but these two, like Earth and I forget some alien race or something, was we're locked in this, in this conflict, but they were so far apart from each other that to send troops, basically, it would take like a thousand years for the troops to arrive, for them to fight right. And you know, long story short, basically, at at you know some point. I think it was near the end of the of the of the novel. You know, the ships they had sent, the ships that they had sent out a thousand years ago were slower than the ships that got sent out 500 years ago so the ones that were sent out afterward were actually arriving before they were because the tech was so much better right like the tech was better and everything.

Tim: 14:39

So I think I kind of wonder if it's like, if it's like that, like we're going to pledge 100 billion dollars to grow this ai infrastructure, but but the tech is getting good enough that maybe we don't need it.

Chris: 14:48

Your take is much cooler in mind because it has a sci-fi background. Where my mind went, I was, like, is this kind of just like the evolution of the data center? To me it's like for a while people were building out their own data centers in a facility, something like that. Now we've kind of got to the point where people don't build out data centers, right like, um, you know, in a facility, something like that, and you know, now we've kind of got to the point where people don't build out data centers. Maybe they rent a rack and a colo or something like that, or maybe they just put it all in cloud. Now, you know, we kind of started out this conversation with like, oh, you need to build your own llm, you need to do all this stuff. It's almost like people like before the, you know, before the boots were ever on the ground with that, you know those marching orders.

Tim: 15:26

We've already skipped it. We've already skipped it.

Chris: 15:29

So it's fucking. Yeah, it's wild. I don't know if that just speaks to kind of like the agility of the market or what, but yeah, man, things are changing on a fucking dime man.

Tim: 15:39

Yeah, and I don't think, and that's why it's so hard to invest. That's what I guess. That's my point right. Amazon's going to invest $100 billion and Google $70 billion. I don't think they really have any idea what that money's for, because the tech is changing so quickly. What does that even mean? By the time you've half built this thing, you spend half the money and the whole freaking technology's changed and now you got to do something completely different, probably.

Chris: 16:02

Hopefully all these orders are processed before the DeepSeek thing, so that the NVIDIA sales reps can surely get their money right.

Tim: 16:11

Yeah, get their commission checks for the quarter, okay. So the next one actually is from CRN in Australia, and this one is an article about how Amazon, the parent company I don't know if parent company is correct, but like I don't know, I don't know, I don't remember what the what the relationship is anymore between Amazon and AWS. Is it parent company? Is it sister, like I don't know how they anyway. So Amazon itself was able to beat Wall Street estimates. So, from a quarterly revenue perspective, so they strong real retail business.

Tim: 16:48

You know, we'll see how long that lasts, but basically, this article from CRN points out that the cloud computing unit of Amazon, aws, reported a 19% rise in revenue to. This is ridiculous to $28.79 billion, falling short of the estimates of $28.87 billion according to the data in this article. So, yeah, so cloud growth lags is what the title says, and to me we're working on such ridiculous numbers. Still, here we are, so many years later, this thing is still growing 19% in a quarter and they missed their quote-unquote target. By what is that? Is that eight million dollars? I don't, what's? 0.87?

Tim: 17:32

like 0.79 and 0.87 billion. Um, you know it's such a ridiculous anyway. Um, targets are, honestly, I'd say that was a pretty good target. Whoever set that target obviously did, you know, was pretty pretty spot on with their, with their estimations, I would think so. Um, and then the article goes on to point out that, like the retail sector offset the aws's quote-unquote cloud weakness which I'm still like scratching my head on, like I guess it'd be, you know, beauty's in the eye of the beholder, or like math is right, I don't know whatever. Um, yeah, so so cloud weakness, cloud growth, lagging at only 19 guys, I'm sorry it's.

Chris: 18:13

We probably should go ahead and repatriate everything at this point yeah, dude, like not much to add here, but this is the kind of shit that just pisses me off. I feel like we're constantly talking about the same numbers in a positive light and then a negative light, and it's like we had one article just talking about their investments in ai and it's like, well, their revenues increased 90 year over year in q4. And then the next article I'm like, oh, that sounds pretty fucking good. Like 19, like that's, you know, as a growth company, that's that's pretty damn good. And then the next one I was like, oh, yeah, that, yeah, that was pretty bad, like we didn't live up to expectations.

Tim: 18:46

Had a terrible quarter. It's like fuck man dude Like Jesus.

Tim: 18:49

Christ. Yeah, I mean we always try to cover the numbers because they usually help us understand a little bit about just the industry in general, where the cloud industry is going and whatnot. But it's tough when you get these article titles that I guess they turn on a dime right, depending on what they want to say. It's either bad or it's good. Personally, I'm of the opinion that 19% growth, like you pointed out, is pretty fucking okay. Like comparatively Sounds pretty nice, considering that AWS is already the 800-pound gorilla in the market.

Chris: 19:24

Exactly that's the thing when they stand today. 19% sounds fucking huge yeah, exactly so.

Tim: 19:31

I mean cisco would love to see some 19 growth. I'm pretty sure right or pick a pick a dominator of the industry that they're in, and I'm sure they'd love to see almost 20 growth. So no big deal here, just kind of a funny, funny article. But uh, yeah, let's move on sweet, all right.

Chris: 19:48

Um. So last one we have today, which, I'll be honest, this um, this obvious, as a excuse me, as a cloud focused podcast, and, uh, this isn't necessarily a name that we hear a lot in the cloud world today, but, um, we have an article from CRN again, um, about solar winds. Um, which, if you're familiar, is the observability and IT management vendor that we've probably all used for many years in the past is being acquired by a San Francisco-based private equity firm called Turn River Capital. Um, companies getting acquired by private equity is usually kind of, uh, a troublesome thing and kind of sad, but, um, you know they, they, of course, spin this as a positive thing. You know the the, the private equity firm is going to of SolarWinds, since, like it was during COVID, when they had the supply chain attack, which was really bad.

Chris: 20:55

Albeit, there was like some dude I'm misremembering here. Was it Russian backhackers or was it Chinese?

Tim: 21:02

I can't remember. I don't remember if it was Russian or Chinese. To be honest, yeah. But it was in their software right. It was a supply chain of one of the packages. So yeah, it was bad.

Chris: 21:10

Yeah. So um, yeah, I don't know if this is kind of the start of the end for for SolarWinds I hope not Um, just cause I know they're. They're a name that has been um relative to me for many years and you know, in terms of network monitoring and things like that. But um, yeah, so um didn't expect to see this. It was a 4.4 billion all cash deal as well, which I thought that sounded rather low um, just because of like how big I assume they still were in the um on-prem world.

Tim: 21:41

But maybe, maybe that's not the case nowadays, but I don't know well, I mean all going, all cash usually gets you a discount so you don't have to like worry about financing and stuff like that. So, but yeah, I mean four billion, 4.4 billion. Yeah, I remember when solar winds was a juggernaut and I'm not saying they're not still, I actually don't know um, I mean, we used it at the last time I had an enterprise job. We used it um many, many times, being getting a getting a call at the end of every single quarter by my SolarWinds rep who was desperate to sell me something. And then if you ever had the misfortune of downloading a free trial of a SolarWinds product you had, if you were unfortunate enough to supply a real phone number, you would basically be on the list forever.

Chris: 22:25

They were definitely very aggressive to this day.

Tim: 22:28

Yeah, they were very aggressive. Now, luckily it was a work phone, they were very aggressive and I don't know. I mean SolarWinds has been around for a while. Anybody in the network space knows SolarWinds, I am I don't know.

Chris: 22:40

Man, I have yet to find the company, the private equity firm, the company bought by a private equity firm that, like you know, pes were good ideas and good companies usually go to die. Yeah, I mean, I'm looking at. I'm looking at Turn River's portfolio as well on their, on their website, and I'll say, of of all the companies that I was seeing here, many of them have exited at some point. That doesn't necessarily say they exited and went on to do great things, but of all the logos on here, I recognize one exited at some point. That doesn't necessarily say they exited and went on to do great things, but of all the logos on here, I recognize one.

Chris: 23:11

So I don't know if that says anything about what this means for SolarWinds. Or maybe I'm just, you know, ignorant to to this market. I mean, there's a lot, of, a lot of. You know, like application stack specific companies in here that I have no business knowing who or what they do, to be honest with you. So that could also be a piece of it. But yeah, I'm not. This doesn't leave me feeling warm and fuzzy about SolarWinds. I'll say that I mean, yeah.

Tim: 23:38

So the whole thing about private equity right is that they acquire companies with the goal of essentially making more money than they paid for the company. Right, if they destroy a company but make you know $10 million off of the process, then they consider that a success and they move on to the next company.

Tim: 23:58

They're the trademark of inshitification, you know. So generally they will buy a company, inshitify it as much as possible as legally possible to do so or raise the prices and just ride out the wait until all the customers leave. You know like they get as much money as they can out of it until the whole thing's run into the ground. I you know. Prove me, feel free to prove me wrong. I'm sure there's some PE out there that didn't destroy the company that they bought. I can't think of any at the moment. I PE out there that didn't destroy the company that they bought? I can't think of any at the moment. I know several, but yeah, so I guess we'll see, time will tell.

Tim: 24:31

On the SolarWinds thing, as always, yeah, time will tell. All right, well, I think that's all the stories we have for this week. If we missed anything, of course, please do let us know in the comments of our YouTube channel or email, or smoke signal, I don't know, discord, however, you want to let us know, that's fine. We'd love to hear from you. We always love to hear from you. Okay, so, on the behalf of myself and Mr Miles, we'll go ahead and close out this episode of the Capels Clouds News Podcast, please follow us Buy our book.

Chris: 25:04

Please buy our book.

Tim: 25:08

Oh yeah, we need to do an episode about the book, probably just to show it off to everybody and get Steve over here, probably should. We'll do that and you guys can see all the cool stuff we've been working on for the last what god like almost two years now. I don't know how long it's been. Anyway, it's been a while Anyway, alright, so look for that in an upcoming episode.

Chris: 25:29

Until then, everyone, take care and see you later. Hi everyone, it's Chris and this has been the Cables to Clouds podcast. Thanks for tuning in today. If you enjoyed our show, please subscribe to us in your favorite podcatcher, as well as subscribe and turn on notifications for our YouTube channel to be notified of all our new episodes. Follow us on socials at Cables to Clouds. You can also visit our website for all of the show notes at CablesToCloudscom. Thanks again for listening and see you next time.