Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this webcast. The author does not hold any equity positions with any company mentioned in this webcast.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.

Transcript:

Krista Macomber: Hello and welcome to episode 22 of Infrastructure Matters. My name is Krista Macomber. I’m joined here fresh off the road from AWS re:Invent as Camberely Bates and Steven Dickens, my co-hosts. Camberely and Steven, how are you feeling this morning?

Steven Dickens: I don’t know whether I’d describe it as fresh off the road. I think a little weary and stumbling back into the office is probably a bit more accurate. Camberely looks fantastic as always, but I’m a little bit jet-lagged and there’s a little bit of struggling this morning.

Krista Macomber: Yep.

Camberely Bates: I’m going to just ditto with him, because I feel like I’ve been run over by a truck. I got the license plate of the truck.

Krista Macomber: Oh, my goodness. It’s-

Camberely Bates: It’s called AWS 123.

Krista Macomber: Yes.

Steven Dickens: Yeah.

Krista Macomber: Yep, yep.

Steven Dickens: Fantastic.

Krista Macomber: Always a sign of a very good productive week and we’ll get to that in just a second. Before we do, I did want to call out, so the Infrastructure Matters podcast this past week, while Camberely and Steven were busy at AWS re:invent, we actually were ranked number two in the top eight podcast for cloud computing via Techopedia, number two. We wanted to do just a quick call out and a thank you to that recognition. We’re very excited and we’ll be sure to link to that article here in our show notes. So we wanted to jump right into our discussion of the show and to kick it off, I know there was, again, a ton of buzz around the show. There’s actually over 60,000 attendees, Camberely and Steven, I know we were just talking about that off camera before we jumped in, a very active show floor. It looks like, based on the photos, certainly very much getting back to it actually far surpassing the pre-COVID levels. I know that AWS itself had over 250 announcements at the show, and of course, it was packed with partners having a bunch of announcements as well. I know I was just referencing that very packed floor. For The Futurum Group and our team, I know between us and The Six Five, actually, we recorded 17 videos. Several of which were actually from the beautiful top 60th floor of the Wynn Hotel and Casino. Very, very busy week for our team as well and I know we wanted to give a big thank you to our production team and, of course, all of the analysts including yourselves that were there. Camberely-

Steven Dickens: Yeah. I echo the comments on the production team, round of applause to those guys. 17 videos in three days is just… Yeah, it’s impressive.

Krista Macomber: Absolutely, absolutely tremendous. With that, I’d like to turn it over to you, Camberely and Steven. You were obviously there, feet on the street all the week. What are some of your perceptions coming out of the show?

Camberely Bates: First of all, we kicked off with an Analysts Days. They flew us in for Sunday, which was a sports day, and taking us through a lot of the work that they’re doing. Amazing work is being done with the sports facility. Then they took us over to the NFL game that was over at the Allegiant, which was great. There’s a lot of people that saw Mark Allen came by the booth where we were and is like, “Wow, okay, so I got to see him.” I think one of the things that I saw there was the amazing work that’s being done with data analytics and data analysis. I think we do know there’s a lot of chips that are in the puck, in the football, and all that. They’re gathering the data in, but what you don’t know is that they’re putting chips on the pads and what they’re wearing, and that data is going into being all gathered in. You can imagine that information that’s being used by the different sports teams and their analysis there, that just gives you some ideas. They also had the soccers and the golf, and those areas. The second day, on Monday, we were there with all the analysts and we kicked off with an analyst day before we even got into the keynotes that were happening. They did a bunch of pre-announcements, laid out what was going to be happening, and getting us ready to march through the next few days, which was, as we already mentioned, pretty exhausting in terms of our measurement of how many miles. I think I logged about 30 miles walking around, that does not count the workout.

Steven Dickens: Good for the the steps. Let’s just leave it at that.

Camberely Bates: Yeah, those are the steps. With that, what Steve and I would like to talk about is some of the big, big themes that were there. Probably the biggest theme was gen AI or AI, period, and what they rolled out. Steve’s going to talk about that, but the other two big ones that I took away from there is, one, this is my theme. They said this, “Hybrid is here.” They acknowledge that hybrid is here and they didn’t fight. It was like, “Okay, this is here, we’re going to deal with it.” That meant that cloud migration, cost optimization, all those items was here as well.

Steven Dickens: Yeah. The big thing that came out of the show for me, there was a couple of big announcements. Q, which Matt Wood talked us through on the pre-brief session that Camberely was talking about. The interesting thing for me, before we get into it, is why it was called “Q,” after the quartermaster in the 007, James Bond movies. I thought that was… It encapsulated for me everything about what they’re trying to achieve with Q, that helpful… Can help you out with whether you need whatever you need in order to get your job done, Q is going to be there for you. I thought that was a great reference. Great throwaway line from Matt, that Amazon owns MGM Studios, so they didn’t have any issues with the trademark, which I thought was fantastic. No, all joking aside, I think the other key takeaway for me with Q was Amazon and AWS, they’re product-led businesses. Two pizza teams, go get a service launched, 250 announcements this week.

He acknowledged this when he was talking us through it, that they’ve taken a more thoughtful approach and not put copilots or AI assist, or whatever you want to call it, into every product, which would’ve been a natural thing for, I think, AWS to have done given how they’re structured. It was more thoughtful, it was above the product level, beyond the service level, and really trying to put a overall position in place with Q. Interesting for the developer, interesting for the cloud admin, I think that’s going to get widely deployed and widely used. My only concern with AWS is just the sheer volume of services, and Adam Selipsky talked about this, 60% more services than their next biggest competitor. That’s obviously fantastic, it’s obviously great, there’s a lot of optionality there. I’ve heard other people talking about, “There’s 28 different ways to run Kubernetes on AWS.” That’s, frankly, daunting for a lot of the admins, so something like Q to be able to simplify, make it easier, make it more easy to navigate. The other big announcement for me was around their custom silicon updates to Graviton for generation four and Trainium version two. I think, with everybody now in the hyperscale-up space, having their own custom silicon, they were very keen to say, “We’ve been at this for a while.” 2018 was when they launched Graviton, it’s on generation four, so there was some, in the closed rooms, a bit more bobbed comments around, “We’ve been at this, we’re not just talking about it, it’s been in production.” I think it was 10,000 clients now running Graviton workloads, that’s impressive for me. I think that was the other key takeaway, there’s a research note that Daniel, Rafa, and myself have published on that. If you want more details, take a look there.

Camberely Bates: To that end, one of the things that did come out as well is their recognition of the complexity was going on. They are starting to address, they didn’t exactly use these words, but I’ll use them, solutions approach. You’ll start to see maybe how they’re… Especially how they rolled out the AI pieces of it, they showed it as a layered cake. Starting out with a base infrastructure, the next piece of it is the data management piece of it, the next piece of it is the systems environment, and the last piece of it is the application area. They are recognizing that. One of their key competitors being Microsoft, they’re well-known for packaging certain things and bringing out packaging. Google had done the same thing with their analytics space, it’s more packaging. Whereas the way that Amazon has gone to market is this very much so, as you said, the two pizza product teams going out there, which is why you have 250 announcements, because you’ve got 250 teams working on whatever the next generation press releases that they’re going to be putting out.

Steven Dickens: Yeah, it was interesting, there was a chart that Adam put up in his keynote. 60% more services and 40% more features than our next competition. Now, he obviously positioned that as really good, but there’s some dissonance for me between 60% more services and only 40% more features. Is that too many services? Then the other point that resonated for me is Microsoft’s doing fantastically with their public cloud, Google the same, do you really need 60% more services to be the number one arguably in the space? Is that complexity? We talked about it in one of the database and data layer, they put up a logo chart and a services chart, and I’ve took pictures of those just because I need to go back and do some work. There’s a lot of options, maybe too many options. All of the partners were there, all of the AWS native services were there. It was an eye chart to consume and I just wonder, there was a gentleman, I’m blanking on his name, who’s going to be picking up a role around verticals. I don’t know whether you were in that briefing with me, Camberely, but it’s going to be interesting to see how they evolve and go forward with packaging, as you mentioned, vertical focus, more curation of services. That was one of the things that was rolling through my mind through the entire event.

Camberely Bates: The beauty of that, also, is that spurs the industry partners and the need for systems integrators, for GSIs, for partners to help clients get through, “What do I need to do?” That also plays into some of the things that Vogel talked about yesterday. He did a presentation on… I think it was the frugal architect, I think is what it was called.

Steven Dickens: Yeah.

Camberely Bates: Talking about architecting. Once you’re up there, really architecting for your cost-effectiveness, cost optimization, performance, et cetera. There’s this huge skill and opportunity for people to step into with that. With complexity, there’s opportunity and also flexibility that goes there. Maybe we should, Krista, go into some of the announcements that were going on just to highlight some of the big stuff, big areas that each of us trapes after and take a look at. I know you mentioned the processors, Steve, a couple of areas that… One of the biggest ones that they headed off with. I love it. Adam heads off with storage, that was the first one he-

Steven Dickens: I thought of you when he went there. I’m sitting two seats away from Camberely and I’m thinking, “She’s going to love this.”

Camberely Bates: He knows where his money is, because that makes the storage… The data storage makes a lot of money for them in profitability, but anyway, the big one was S3 Express One Zone, roll that off your tongue. I have to look it up every time I have to say it, because the name is not memorable. What it is memorable, though. What they’ve done is… What we’ve seen is that object storage, or S3 if you want to call it that, is going into a high performance mode. It’s serving new areas that have to do high performance. Computing, gaming, et cetera, so this is an offering to address that market. Specifically what they’ve done is they’ve taken the S3, put it into 1AZ, which reduces the latency that you have when you have the capabilities that it has. Now we’re now down into millisecond latency. It’s 10 times faster than the S3 standard, and that means that it’s opening up the door for data protection, et cetera, because you need to have data protection. You saw Haiku, Veritas, Commvault announced with it as well as Vicinity, which is another company which is really interesting that is going to be helping them move the data into S3, because there’s only four AZs right now that this is sitting in. Talking to the folks that are involved with this, there are some really big companies that are looking at this, because it does address some major issues that you need to have addressed within terms of performance within the S3 space. With that, I’ll turn it over to my friend, Krista, talk about her stuff.

Krista Macomber: Sure, sure. Remotely, I did actually cover Camberely. They had a couple of new features on the data protection front and, from my perspective, there’s very much of a parallel in terms of what we were just commenting on, from a macro level for AWS. Steven and Camberely, as you were talking, I was nodding my head a lot and really thinking a lot of signs in terms of pointing to AWS really taking that next step in terms of becoming that even more mature enterprise-grade player. I think, of course, we’re seeing… Obviously, a lot of enterprises are relying on a variety of AWS services, but trying to reduce that complexity and package things up to be more consumable, I think that’s an important next step for them in addition to continuing to fill out those feature capabilities. For me, that was a story that we’re seeing in the data protection space. A couple of things that jumped out from that perspective to me were… For AWS backup, they added the ability to automatically test recoveries, and similarly with the AWS disaster recovery service, DRS, the ability to automate the validation of EC2 instances that are used for recovery as well as things like network connectivity and the availability of free storage space.

Looking at where AWS has been sitting in data protection, I’m not necessarily sold yet that they’re going to go ahead and try to compete against some of these partners that you were just mentioning a couple of minutes ago, Camberely, the likes of a Haiku or Rubrik and some of those third-party backup and data protection providers, but where it is making sense and where their customer base may be asking for it beginning to fill out some of those capabilities. From my perspective, that was really the takeaway there. Specifically when we think about validating and testing recovery capabilities, when we look at the data protection market, that’s becoming more important these days, because of course, we’re seeing the onslaught of cyber attacks, so organizations are facing the pressure to prove that they can recover, and that’s from the standpoint of different legislation that they need to comply with. It’s also really just pressure from the business. IT operations teams, whether or not they’re in a highly regulated industry or region, they’re getting asked by the business to be able to say, “Hey, okay. If we are faced with one of these cyber attacks, we can hit these certain recovery times and we can make sure that we’re going to achieve this level of minimizing data loss. Those were two of the bigger things that jumped out for me from the data protection space and certainly some parallels with that macro level view there.

Steven Dickens: Yeah. For me, over and above the Q and the Graviton and Trainium chip announcement, the other one in my space that came out… There was a joint collaboration with Kyndryl. I got pre-briefed on that prior to the event and commented in their press release announcement, the AWSM2, their managed mainframe migration service. Kyndryl has got the probably the… Not “Probably,” definitely the largest mainframe outsourced installed base with over 6 million MIPS under management. They’ve got probably the biggest bench or maybe even more than probably, definitely the biggest bench of mainframe skills, but they’re going further than that and have connected in the Kyndryl bridge technology, which is how you manage Kyndryl services. There’s a real deep technical integration with Kyndryl, gets super techie and super nerdy. Maybe we’ll link to that in the show notes, so you can read our coverage. Me and Rafa have put some coverage together on that, but I met with a number of different vendors, we recorded some video with Neil Fowler. I think AWS is really leaning into that mainframe modernization space and providing optionality. The other one that, on a collaboration piece while I’m talking about IBM, was that DB2 now runs on as a service on AWS. I think if you’d have said that to anybody five years ago, they would’ve just spat their drink out and just laughed and called you crazy.

Camberely Bates: This is giving to that first thing, hybrid is here. Cloud migration, right?

Krista Macomber: Yeah.

Camberely Bates: What they said is they’re meeting where the customers are, which is this is where they’re at. If they want to engage in this, you’re going to have to say… I have to look at their environment and, no, they’re not going to convert the DB2. It’s just not going to happen.

Steven Dickens: Yeah. They gave us an outpost update as well. I think the level of pragmatism is very different to where it was five years ago. Hybrid platforms like DB2, customers have got massive investments there and they’re going to keep them. The answer is not just move it to a new AWS-native database service and just throw everything away as you move. People are going to be more pragmatic than that, I think.

Camberely Bates: Yeah. Absolutely. Within that, we had a Q and A with Adam, the CEO, after some of the keynotes, et cetera, for about an hour. One of the questions was… There was a couple of the different questions were asked that were along these lines, which… What is the tailwind for the company? What is the headwind for the company? Et cetera. The tailwind is pretty much obviously for everybody, and we can see with the earnings and everything else that came out this week, is the AI and how rapidly that is going. It’s not just lean, mean startup companies, we had a presentation from Travelers Insurance, 160-year-old company that is doing some amazing, amazing things in implementation and they’re doing it jointly with AWS, which means they’re doing it in the cloud. Take that one on. You know they have got a mainframe someplace back there. The second one is what the headwind is. He talked about there’s a general uncertainty in the industry or, actually, in the economy for everybody, I think there’s uncertainty.

There’s so much dynamics going on within the geopolitical world as well as our own nation that there is this cautiousness. Where that builds into the strategy is this cost optimization and, how do we really work with our customers to look at? How do we make sure that they’re really using the systems and taking advantage of the systems? We saw on stage, actually, with Werner’s presentation yesterday as well, people talking about what they had done strategically and looked at it. One of the big names that I’ll call out is Newbank, the born-in-the-cloud bank that’s servicing countries like Brazil, and how they have gone through but redesigned within a point that I think about… This is how our architects on premise redesigned. You’re looking at everything from the code to the networking, the entire piece of it, looking at… It’s the same skill base. Similar skill base, it’s not the same. It’s a similar skill base to what was there. We’ve had on-prem, but we’ve now lifted it to another area, so it doesn’t disappear. It’s not like they are disappearing here, they’re really, really needing the folks up there. That was another piece I’ll bring out.

Krista Macomber: Certainly.

Steven Dickens: Busy week, busy week. We could record this for the next three hours.

Krista Macomber: I know.

Camberely Bates: I feel like I’ve just scratched the surface on here about what’s been rolled out.

Krista Macomber: I know. As we mentioned throughout the discussion, plenty of content coming out. We had a number of videos that were recorded, number of research notes, so the content engine is going to be continuing to be firing on all cylinders here coming out of AWS. Because it was not a busy enough week or time period since we last recorded, in our last couple of minutes here, we did want to touch on the close of the acquisition of VMWare by Broadcom. Obviously, some pretty big ramifications in the industry. I know we’ve gotten some details in terms of how things are going to shake out. I believe VMWare is going to continue to operate as four distinct units. One for cloud foundation, one for Tanzu, one for the software-defined edge and application space, and another for networking and security. With that, Steve and Camberely, anything you wanted to comment on regarding that before we close out here for this session?

Steven Dickens: Yeah. Obviously, the press is full of a lot of the layoffs that have been announced.

Krista Macomber: Yeah.

Steven Dickens: Obviously, that’s horrible for individuals involved, especially the time of year, but I think it’s no surprise to anybody. We’ve been seeing this play out. The team at Broadcom do a very, very good job of running a flat organization. We’re really exposed to some of their team. Hock is hands-on, there’s no layers between him and the business unit leaders. They do a very good job of the efficiency of the backend, so the HR, the finance functions, which is fantastic for shareholders and you’ve seen that with Broadcom’s share price over the last year. I think, yes, it’s horrible on a personal level, but I think it’s not surprising. I think the structure that you just mentioned and the four business units, Raghu’s going to stay on as a technical advisor, Sumit, who was the president there, looks to be moving on, which is a shame. I was impressed with him. Obviously, things change and people move on. If you want to see the coverage, Daniel and Patrick got time to spend with Hock to really drill down on where he sees the vision for the company. That was part of the announcement on Day of Close. I think, the way I look at it, yes, there’s going to be some not great headlines in the first few weeks of this, but I’m super bullish long-term Hock and the team have briefed us, 2 billion of investment coming in. I think VMware wasn’t a bed of roses before the acquisition, it was growing at single digits. Given their portfolio, it should be double-digit growth, not single digit growth, so I think streamlining some of the product portfolio… Camberely, keep me honest, was it 120,000 SKUs that they told us they had?

Camberely Bates: I can’t remember, but it was a lot. Ridiculous.

Steven Dickens: That’s bloat that needs to be trimmed down, so I think what you’ll see is a more streamlined, more focused, more invested-in slicker business that pops out of this. Maybe three months from now, being super pragmatic. It’s going to take a while for the organization to digest some of the personnel changes and people to understand where they fit. Maybe not the super short term, but the short term and the medium term, and definitely the long term, I’m bullish.

Camberely Bates: Absolutely.

Krista Macomber: Absolutely. All right, with that, I am very impressed that we were able to keep this within our 30 minutes here, but I think-

Steven Dickens: I’m impressed we’re able to stay awake, basically.

Krista Macomber: Certainly.

Camberely Bates: It’s going to be one of those ways we get off and go, “We should have mentioned that.”

Krista Macomber: Right, right.

Camberely Bates: What about that? I’m going to say to all my buddies out there-.

Steven Dickens: 250, yeah.

Krista Macomber: Yep, yep, yep.

Camberely Bates: Sorry I didn’t call out to you, too. There we go.

Steven Dickens: Every product manager that we ignored in this podcast, we apologize. You did great work.

Krista Macomber: Absolutely. All right. With that, Camberely, Steven, thank you so much. I’m going to let you guys go and hopefully get some rest, as we look to head into this weekend. We wanted to thank everybody for joining us, episode 22 already. As always, please make sure that you like and subscribe to make sure we don’t miss any future episodes. I know it’s getting into the holiday season here in the United States, but still a ton of travel events and plenty to keep us rolling into the new year, so we look forward to seeing you on the next one. Thank you all so much.

Camberely Bates: Have a good one.

Other Insights from The Futurum Group:

AWS and Kyndryl Partner to Build an AI Innovation Factory

AWS MadPot Honeypot Thwarts Cyberattacks from Nation-State Actors

AWS Enhances Disaster Recovery with Automated Testing and Validation