How is AI Affecting CSP Earnings? - NC2C022 - podcast episode cover

How is AI Affecting CSP Earnings? - NC2C022

Nov 06, 202426 minEp. 22
--:--
--:--
Download Metacast podcast app
Listen to this episode in Metacast mobile app
Don't just listen to podcasts. Learn from them with transcripts, summaries, and chapters for every episode. Skim, search, and bookmark insights. Learn more

Episode description

Send us a text

Is cloud computing just the tip of the iceberg in the tech industry's financial landscape? Join us on Cables2Clouds as we uncover the impressive earnings of Amazon, Microsoft, and Google in their cloud divisions. Amazon's AWS has reached new heights with $26 billion in revenue, while Microsoft's robust performance in cloud computing stands resilient despite bumps in other areas. Meanwhile, Google's cloud business is riding the AI wave, boasting a staggering 33-35% growth. With insights from analyst Dan Ives, we venture beyond the giants to explore the rising curiosity around smaller players like Oracle and Alibaba, and their potential role in AI workloads.

As AI's influence grows, so does its impact on cloud networking. We explore how generative AI is not just a buzzword but a transformative force outpacing even cloud computing's evolution. Platforms like ChatGPT have changed the way society engages with AI, presenting new challenges and opportunities. We delve into AWS's CEO's claim of generative AI's rapid expansion and the potential strain on resources it may usher. On the security front, innovations such as distributed cloud firewalls, Zero Trust Network Access, and multi-cloud connectivity are reshaping how enterprises safeguard their digital environments. Tune in to discover how these advancements are setting the stage for the future of cloud and AI.

Purchase Chris and Tim's new book on AWS Cloud Networking: https://www.amazon.com/Certified-Advanced-Networking-Certification-certification/dp/1835080839/

Check out the Fortnightly Cloud Networking News
https://docs.google.com/document/d/1fkBWCGwXDUX9OfZ9_MvSVup8tJJzJeqrauaE6VPT2b0/

Visit our website and subscribe: https://www.cables2clouds.com/
Follow us on BlueSky: https://bsky.app/profile/cables2clouds.com
Follow us on YouTube: https://www.youtube.com/@cables2clouds/
Follow us on TikTok: https://www.tiktok.com/@cables2clouds
Merch Store: https://store.cables2clouds.com/
Join the Discord Study group: https://artofneteng.com/iaatj

Transcript

Cloud Earnings and AI Monetization

Speaker 1

Welcome to the Cables to Clouds podcast , your one-stop shop for all things hybrid and multi-cloud networking Now here are your hosts , Tim , Chris and Alex .

Speaker 2

Hello and welcome back to another fortnightly edition of the Cables to Clouds news podcast . So thanks for joining in today .

My name is Chris Miles at BGP main on Twitter , as always with me my beautiful , beautiful co-host , Tim McConaughey at Juan Golbez on Twitter , and we are here to bring you some of the news that's dropped over the last few weeks in cloud and cloud networking and things like that .

I will say it is earnings week here on the new show today , so we're going to be talking about some earnings and then a couple other announcements that we found to be interesting . So , first up , we have kind of a couple of articles that have covered all the earnings from the big three . So we have Amazon , Microsoft and Google in play here .

So we have articles in the show notes from the Guardian , Reuters and CNBC covering all this , but we'll kind of get into what that looks like overall . So Amazon seems like they did probably the best , I would think , in terms of overall growth or earnings they came out about .

I think it was like $1.7 billion above the expected earnings according to Bloomberg , and specifically within Amazon Web Services , they pulled in about $26 billion in revenue compared to $22 billion for the same period last year which is obviously good .

Microsoft , oddly enough , also came in better than expected , at least within the cloud computing division , but still suffered in other areas , causing their stock price to drop a little bit . And then you know , I guess I guess specifically in cloud I should say Google probably did the best . They saw a 33 percent increase or 35 percent . It's the one .

Article says one thing , the other says another thing , so we'll call it . Somewhere in the middle there Grew about 30 , 34 , 35 percent , all relating to the , the AI boom obviously , which is to be expected .

One one call out that I saw in here that I really thought was where my mind was headed anyway , where my line was headed anyway A comment from Dan Ives , who's a Wedbush analyst , saying that investors need to see the monetization of AI spreading to the rest of the tech landscape , and the next few weeks will be the linchpin to confirmation that the AI use case

phase have now begun within the enterprise world , and I think that's like I said . We're coming up on conference season as well , so I'm expecting something's going to have to drop .

There's going to have to be something that is maybe not the ultimate game changer , but there's got to be something in that realm that is going to prove how AI is going to change specifically the cloud computing landscape in some capacity .

What that's going to be , I don't really know , but I feel like we have to get something out of this , or people are going to start pulling out in some capacity , at least from my perspective . I don't know . How do you feel , Tim ?

Speaker 1

Yeah , I mean there's a lot of bearing the lead in a lot of these numbers , I think . So , for example , I'll just jump on the Google number 33% growth . I think Google is still a little bit in lagging numbers wise , so 33% growth is really good for them .

Speaker 2

But I think it's all relative .

Speaker 1

Exactly that's what I was going to say . You have kind of a law of small numbers , law of large numbers thing where , once you get to a certain point , moving the needle at all becomes our killing effort and the , the percentage is less useful than , like you know , absolute value numbers . But that is good for them .

And uh , I'm , I'm , you know , glad to see that gcp or google cloud sorry is uh is seeing that uptick in growth as well . Uh , I would like to .

Actually , what I'd love to see and it's not part of this but because the articles weren't put out about the other like Oracle and Alibaba and some of those smaller clouds , I'm curious if they're also seeing , especially Oracle , that's like subletting compute to OpenAI and stuff like that for AI workloads , if they're seeing an equal or at least commensurate amount of

growth there . So , like that was , that was one thing that , you know , kind of stuck in my mind . I couldn't agree more that with this guy saying that we need to see the monetization of AI go beyond kind of the first party , if you will , where the CSPs are the ones seeing the money , if you will from , from offering that CSP or that AI service .

You know , if you look at the AI services , a lot of the managed services have specifically to do with models accessible . And then you know the customers , which are businesses that are ostensibly going to do something with that model , with that managed service , are kind of left to their own devices to kind of create value , a valuable product , using the Gen AI .

And I think everybody's kind of left to their own devices to kind of create value , a valuable product , using the gen AI . And I think everybody's kind of treading water , or more like flailing around in the kiddie pool without the water wings on , trying to figure out what does monetization of it .

Everybody's building an AI chatbot , for example , like that's oh , look open AI , build an AI chatbot . And that's the lens now through which everybody looks at AI , you know , to some degree as as useful and I think a chatbot's cool , but I think it's not even scratching the surface of what the actual technology should be used for .

Speaker 2

So that's yeah , the chatbot seems like it's the easy insertion point . Right , it's the easy way to get your hands dirty and then take it somewhere from there . But I think people are expecting hopefully , yeah , hopefully , um . But also I think there's some people out there expecting like oh well , we made this chat bot and it does fuck all .

So you know , ai is not for us type thing , but it's like yeah , you know .

Speaker 1

No , I completely agree . There's a lot of . I think what we're going to see is what we've already been seeing , which is like what was that ? Uh , ai assistant the , the ? What was the guy that reviewed it and said it was terrible ? You don't remember the AI ?

Speaker 2

assistant that could project MBK or MKB . What was ?

Speaker 1

that ? Yeah , yeah , yeah , he reviewed it . But what was it called ? Do you remember ?

Speaker 2

I can't remember now the little pin thing , right , what was it called ? Yeah , yeah , yeah .

Speaker 1

Yeah , whatever , it was supposed to be an AI assistant and it was just not ready , not even by half right , but I got to have to say at least the ambition was there to do interesting things beyond a chatbot . So I think really the next it's almost like a race right , with gates in the middle of the race to see who can get past the gates .

The first gate was make a chatbot , and now I think we're getting to the point where , like , all right , well , the chat bot's cool and all , but let's see actual value pulled out of of what we can actually do here , and I think that's we're going to find a lot of businesses . It's stuck on that first rung of the ladder and don't see value .

So , anyway , not to ramble too much about it . But yeah , that was the wrong word . Anything with the stock market is weird this way , where they make money and yet they lose stock value . It's just like the gestalt of stock market . I guess I'm not sure , but whatever it'll be back to up 5% in a week . That's just the way the stock market moves .

Generally , though , what we're seeing is an increase in money , but I think a lot of the CSPs are seeing also they're scraping the bottom of the barrel on resources , like we're starting to see CSPs are . They can't just keep growing like hand over fist right Like they're . They're going to have resources constraints . A lot of them are already hitting it .

Speaker 2

Yeah , that's a good point . Um , yeah , I think there was a specific call out about Azure where it was like you know , they , they . Well , let me find the exact wording .

Speaker 1

Yeah , I mean they can't . You can't just have infinite growth in the cloud . We've said that multiple times . At some point you run out of capacity or you've oversubscribed to the point where you're DDoSing every other customer Like it's . You know they have to . They can't grow forever .

Speaker 2

Yeah , just , I mean , even the first call out in this Reuters article is , you know , slower growth in its cloud business signaling . The big AI investments were not enough to keep up pace with its capacity constraints at its data center , so it's like that was even before the AI craze came into place , right .

So yeah , I'm sure it's just kind of exponentially growing Exacerbated yeah , not that it really I don't think it really had an effect . But something that is spinning around in my mind is also the whole return to office thing , for Amazon has not started yet , so I wonder if that will have any effect or if it hasn't .

I mean , we've already seen some pretty notable people jump ship at Amazon . Oh , exit , yeah , and they've retired a few products and services . So I don't know if that's led into anything or if that was obviously timed in that way , but worth keeping note . I think that you know .

I think their landscape is going to change quite a bit come January 1 , even post-re-invent .

Speaker 1

Oh yeah .

Speaker 2

So , yeah , we'll definitely have to stay glued in for that one .

Speaker 1

Yeah , they've been pretty unapologetically said like take it or leave it . So I am curious to see if there's any kind of brain drain . And also the VP of AI , like earlier in October , departed AWS as well .

So what I find interesting is so the first Gen AI bubble here is CSPs offering build your own , like the , as a service , right , I don't know what you'd call it . Build your own as a service , like not really SaaS , I don't know . I guess SaaS is probably the closest , or PaaS .

Really it's more like PaaS platform as a service , I think right , I'm has yeah platform as a service , I think right , I'm curious to see if not only are the csp's offering this kind of paths model for access , hard models and build whatever you want , but are they going to be doing on the back end , are they going to be

Emerging AI Capabilities in Cloud Networking

in ? You know , we saw this a little bit with q . Remember the last reinvent where they had amazon q with the troubleshooting chatbot or whatever . Again , chatbot is the first thing everybody does , right , and I remember us at the time thinking it was a pretty not ready . It was not ready for primetime .

But it's been a year and , honestly , a chatbot might not be the best way to surface some of that stuff that you want to see . So I wonder if the CSPs or vendors are going to be building other AI capabilities .

Speaker 2

I think that all kind of trickles back in a way to chat GPT right , like GPT had us as a platform society interacting with AI on a at a scale that we never were before , right . So that's like when you first think of AI , it's like I need to talk to it and get some information out of it .

So I feel like that is just the easiest bar of entry , um , or lowest bar of entry to to to kind of give people back what they're already familiar with , right , and just use it in a new um , a new facet , interacting with their own infrastructure and their own applications and things like that . So I mean I can draw the natural line there .

But yeah , you're right , I think it needs to change . But you have to change the way people think about it , right , and that's probably where we're headed , right . They're going to have to introduce a new way of interpreting AI and what JNI means .

Speaker 1

Yeah , and you know , whenever you go back and look at things at disruption , disrupting technologies almost always that disrupting technology like didn't , there was , nobody was thinking about it before it showed up , right , nobody considered what that was before it happened , right , the innovators dilemma , and it's absolutely full of case studies where an incumbent was thrown

out of whack , completely unable to react because some other new technology that nobody even knew was what people would love or that needed or they needed it just showed up on the , on the landscape . So , yeah , I think we haven't seen like . We haven't seen what we haven't seen . It's like almost like a , almost like an AI singularity .

You know , we haven't seen what's what people are doing , what's going to be possible in the next round , if you will , of of creating . So , uh , that sounds a little bit more hopeful , a little than I probably intended . But , yeah , we uh , we'll see , right , um , what people are going to do with it .

Speaker 2

I hope it's just not more fucking chatbots tim's now a glass half full guy on on ai , so that's good , we've changed him .

Speaker 1

ladies , ladies and gentlemen , yeah , totally Speaking of AI , we have an article here from Hold on . I don't have it up right now . We have an article here from Yahoo Finance , right , yahoo Finance saying that the Amazon CEO is saying that Gen AI is growing three times faster than this .

Is AWS's CEO so saying Gen AI is growing three times faster than this is AWS's CEO so saying Gen A is growing three times faster than cloud computing did .

So , assuming that what he's saying is the benchmark is basically how fast the cloud computing , like AWS's cloud computing division , grew , he's saying that just their AI division , or offer or whatever , is growing three times faster . I don't quite know what to make of that except to say , hey , great , ai is growing , like that's obvious to everyone .

Again , I think we're running into , I think we're going to hit resource constraint . Is this the article ? There's an article . One of the articles we looked at and I'm not sure if it's this one , I'll have to bring it up was talking specifically about . You know , was talking specifically about , as CSPs need more and more AI inference .

They're doing their own inference and training chips . At Amazon , they're building their own silicon . So , again , resource constraints , right , so they're building stuff specifically for this . But you're going to run into resource constraints . There's only so many data center rack space . There's only so much silicon out there .

Other people are going to be doing the same thing . Other CSPs are going to be doing the same thing . Other CSPs are going to be doing the same thing . So I'm curious to see how far this can grow before we really start not seeing the growth that investors are expecting because we're hitting hard resource constraints .

Speaker 2

I don't know if I'd call it nuanced , but I feel like it probably is growing faster than cloud computing , just because of the kind of more vast coverage that AI has on the tech world than cloud computing did . I feel like when cloud came into the game , it was more like hey , you already buy this stuff off the shelf , right ?

You're already buying compute , you're buying storage , you're buying networking , right . It was a simple trade-off in someone's mind . But now here it's like okay , if I want to build AI apps , I probably can't buy that off the shelf right now . These are the players that have all of that capacity . They have all the GPUs . I can consume them as a service .

I'm already in their ecosystem because I run AWS , I run Azure , whatever . So I feel like you kind of got to take it with a grain of salt . But at the same time I mean it's substantial , right . I mean we can't go a fucking week without talking about this shit . So it's obviously got some merit , right , but we'll have to see where it goes .

Speaker 1

Yeah , I mean , to me the reason it's growing so fast is because , compared to like original cloud computing division or whatnot , like when cloud first came on the scene , is that , like the train tracks already laid right for for JNI , like the train tracks were not literally laid when cloud computing really started Right , they had very little capacity , they had very

few services . You know they were building the track as they were on it . Here the train tracks are already laid . I mean , like I said , we've talked about this several times today that you know they've already got the bring your own or not bring your own , sorry , build your own app and we're just providing the building blocks for you .

But , like it was very , it was obviously much easier for them to do that part of it than to build out a whole cloud computing infrastructure and fabric and all of that automation . All right , let's see .

Okay , so we're going to move on to a couple cloud networking vendors , specifically cloud networking vendors , because we do still try to once in a while , be a cloud networking-focused podcast as much as we can . Okay , so a couple press releases here .

Usually pretty mean to press releases , but that's because a lot of press releases are extremely what's the word you would use for it . Just like I don't know , just full of themselves Hand-wavy , very hand-wavy . Yeah , very hand-wavy , very full of themselves . These two are not , at least , I didn't think so .

When I read them , I felt they were pretty clear on the benefit or what they're trying to show as a new capability . So we'll go over it . Of course , full , what's the word ? I'm looking for Disclosure , chris , and I do work for one of these companies , so just take that for what it's worth . We're not here to shill any particular vendors .

Speaker 2

We also do not represent the thoughts and things like that of our employees .

Speaker 1

Everything we say does not reflect upon or have anything to do with our employer as well . So for what that's worth , okay , so this first one is a press release from Aviatrix , and this came out a couple days ago , and there's a new version of the Aviatrix cloud networking platform that's been released .

Two of the new capabilities that have been released for it are the hybrid cloud transit and an enhanced distributed cloud firewall . So , really quick , just kind of dig into this . What is hybrid cloud transit thing from the press release specifically , just kind of dig into this . What is hybrid cloud transit thing from the press release specifically ?

Hybrid cloud transit extends the company's secure cloud networking fabric to major data centers and middle mile providers such as Equinix and Megaport . So before this existed , we kind of had data centers and middle mile as kind of an island .

If you will like you could connect to the cloud from the Megaport or Equinix or your own data center and then , inside the cloud , of course , do all the cloud networking .

This is a new capability that actually allows you to use really your own fiber if you have it , or you could use Equinix or Megaport any kind of middle mile provider and use their backbone or your own backbone , just like you would a cloud backbone , basically to move between clouds , to move between regions .

That's just a new capability that moves that kind of cloud networking focus down out of the cloud to you know , middle mile or your own data center . So that's the hybrid cloud

Cloud Security and Networking Innovations

transit . And then a new enhancement to distributed cloud firewall , which is kind of well . Honestly , it's in the name right Distributed cloud firewall , threat intelligence , firewalling , distributed data , plane stuff . They've added a new thing called threat groups and geo blocking via geo groups . This is an extension or enhancement to the distributed cloud firewall .

Basically , it gives you more criteria and more threat intelligence by which you can build security policy .

So you know , for those of us who you know are or used to be firewall jockeys , pretty much any way , any extra ability you have to classify traffic or classify workloads or granularize I think I just coined that word now granularize your security policy , right , that's , that's always good , that's always gold , as long as , of course , it makes sense and it's a

good , useful capability to have . So this adds threat groups and geo-blocking so you can use those along with all the other abilities that it has for your security policy . Let's see what else does it say oh , in addition , this is a big one too . In addition , shibuya Cloud Firewall now supports improved logging capabilities for seamless integration with enterprise SIMs .

So this was something that was a little bit of a that what did exist before , but it was a little harder , a little more manual process . So it's now built into the product for seamless integration with pretty much any of your SIM capabilities and obviously that gives you visualization on the activity and what's happening inside your threat intelligence platform .

So pretty cool capabilities , yeah , and obviously there's more on the way . I could say more , but I will not . I'm going to stick to the press release . So all right , and then let's see . We'll move on to the next one , which is for cloud networking vendor Alkira , and I want to bring up this article because I want to speak directly to the press release .

So yeah , so Alkira announced a new zero trust network access capability for their platform . Of course , ztna is a little bit of a loaded word . There's a little marketing behind it , so you kind of have to really dig in on it and understand what ZTNA means , based on who it is that's saying it , what capabilities they talk about .

So Alkira talks about ZTNA here . If you don't understand how Alkira works already . It might be a little hard , but basically Alkira has these cloud exchange points throughout all the clouds , all the regions where you onboard yourself into the Alkira framework , into the fabric , and then that's kind of ride their backbone .

So I think they say that their infrastructure as a service , a network infrastructure as a service , should I say ? So that's kind of their I don't know tagline , not tagline , that's where they sit , if you will , in the stack .

So the ZTNA ability or capability that's been introduced specifically seems to be focused on kind of the user access , zero trust user access part , so like kind of like a Zscaler or Prisma network access type of connectivity and yeah , and then let me see if I can find the exact thing , okay , so yeah , they go over some key features .

What I think is interesting in the key features they allow you to bring your vendor of choice to secure traffic to internet and private applications through a single security stack .

So I think when I was talking to Will and we've had Will Collins on the show several times , right , and Will of course works for Alkira and I was asking him a lot about how this whole thing works .

So Alkira has kind of a marketplace model , a little bit like the AWS marketplace or the CSP marketplaces in general , where you can kind of subscribe to different vendors . So my understanding is that that's what they're talking about here , so kind of bring your CTNA , andna and use your vendor of choice , but this new capability is built into Alcura .

Any thoughts on either of those ?

Speaker 2

No , I think it's a good combination Because we've talked about even on our last episode that came out last week , talking about concepts of zero trust and things like that and this idea of a kind of a perimeter-less type deployment right . So you know this thought of you know Alkira coming in with their .

You know they have cloud exchange points , so they essentially have a pretty large backbone that you can ride across to build multi-cloud connectivity , allowing remote users to just tap into that and get the inherent security on top of it seems like a good play , right .

Obviously , there's value in that we talk to customers that struggle with that exact thing as well . So , yeah , it totally makes sense . I think this is a good move . Yeah , definitely .

Speaker 1

And I forgot to ask after the last one about the Aviatrix one do you have any opinion on that or anything to add to that one as well ?

Speaker 2

Yeah , I think that one . I got to be careful right Because I work in sales so I feel like I don't want to talk and sound too much like a sales guy . But the HyperCloud transit option is really interesting and really just full disclosure .

Like we said , I work for Aviatrix but we are seeing customers implement some really cool kind of topologies using those middle mob providers like Equinix and Megaport and building , you know , kind of pretty cool backbones across the between cloud providers , between regions etc .

And they're able to get really fine tuned traffic engineering where they were not able to before and it was all kind of do-it-yourself . So it's cool to see the options built into that , with the added visibility and encryption and things like that . So yeah , I'm just saying I'm seeing some people do some really cool stuff , so I'm really excited about that one .

But that's where I'll leave it .

Speaker 1

Yeah , it's fair enough . I mean on the show , obviously everybody has to work somewhere . So you know , the fact that Chris and I work for Aviatrix is this is obviously not an Aviatrix podcast and if we've said many times , we tried not to , we try to walk a fine line and still give fair coverage to to everyone . So just take that before it's worth .

Speaker 2

This is a niche podcast and so you're going to get some . There's going to be a lot of overlap sometimes . All right , great . So with that we will go ahead and wrap up , and thanks for joining us today . We'll be back again in two weeks with more cloud networking news . Given the time , yeah well , Ignite may have happened by then , right , is that right ?

Speaker 1

It's pretty short , it's pretty close . By the next time it's out it'll definitely be over .

Speaker 2

But the next time we record it it might . Fair enough . So we'll keep the suspense and say , over the next few episodes I'm sure there will be quite a bit coming out . You know they always , they always leak some some of the information prior to the conferences and things like that . So maybe we'll get some some early things trickling in and all that .

So with that we'll wrap it up . If you enjoyed this , please like , subscribe , do all the cool stuff , Reach out to Tim , go to his house , give him a hug , all that kind of stuff , please .

Speaker 1

I'm so go to his house , give him a hug all that kind of stuff and uh and uh .

Speaker 2

With that , we'll take it away and we'll uh see you next week . Bye , bye .

Speaker 1

Take care , see ya . Hi everyone , it's Tim 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 podcast catcher , 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 Cables2Clouds .

You can also visit our website for all the show notes at Cables2Cloudscom . Thanks again for listening and see you next time .

Transcript source: Provided by creator in RSS feed: download file
For the best experience, listen in Metacast app for iOS or Android