And that is the opportunity that we are entirely focused on, at least when it comes to campus and branch. Every single one of our access point is connected to the cloud, managed through the cloud with an AIOps engine, Marvis in the cloud. And all of the growth that we’re seeing now in the wired switching as well as in the WAN is happening through the cloud. And even the security capabilities, such as network access control that we’ve introduced, is a cloud-based solution as well. So I think it’s a little bit — it’s wrong to look at the growth of just the campus and branch market all up. It’s actually more useful to look at it from the standpoint of that cloud-connected portion of the market that’s actually growing at a much faster clip. Then in terms of the solution itself, really, when it comes to harnessing the power of AIOps, and I know it’s a bit of a used and abused words out there.
I mean the proof is in the pudding. Our customers today are seeing real benefits. Reduction in number of tickets by 90-plus percent, reduction in total — the time frame it takes to deploy a new solution from what used to be a year to a matter of weeks, if not maybe a month or so. The root cause analysis that used to take days with all of the frustration of people that are trying to use the network to do whatever they want to do has been reduced to essentially no time at all because most of the time, we’re proactively identifying issues and fixing them. You mentioned Zoom. We have already integrated Zoom visibility into our Mist solutions so that when there is, in fact, some sort of a video issue, we can, if not proactively, instantly provide the IT staff visibility into whether the issue is more in the application side, the wireless side, the wired side, the cloud, et cetera.
And that’s the capability that our IT customers are really, really happy to see. So I know I’m maybe beating a little bit of a dead horse here. I’m very optimistic about the competitiveness of our solutions in the market, whether it be a challenged market or not.
Ken Miller: And when it comes to 2024 revenue growth, it’s a little too early to provide guidance for 2024, particularly with some of the weakness we’re seeing with the Cloud and Service Provider customers. But let me walk you through some of the puts and takes. I mean, clearly, the backlog drawdown that we’re going through in 2023 is going to provide a pretty significant headwind to revenue in 2024. Orders are going to need to accelerate just to offset some of that backlog-related headwind that we’re experiencing from a revenue perspective in 2023. With that said, I do expect orders to accelerate in 2024. I expect full year growth across all of our verticals. I also expect Enterprise revenue to grow in 2024 on a full year basis.
So really it comes down to Cloud and SP and the timing and pace of that recovery. And that’s a little bit too early to call at this point. So we’re not giving specific guidance on 2024 other than based on the backlog expectations and our order expectations in Q1, we do expect to see a return to more traditional patterns.
James Fish: Very helpful color guys. Thanks.
Operator: Thank you. The next question is coming from Karl Ackerman from BNP Paribas. Karl, your line is live.
Karl Ackerman: Yes, thank you. I have two questions. Rami, perhaps a question for you to start. So the upside in the quarter appears to be on the services side, while product growth is going the other way. I understand there is a mix dynamic at play on services given the growth that the company is seeing in Enterprise. But where do you think we are in the cycle for Service Provider and Cloud spending on product hardware?
Rami Rahim: Yeah. So Cloud and SP, as I mentioned, they’re definitely going through a period of digestion after they — for the last couple of years, have bought a lot of equipment. I mean, if you recall, in the Cloud Provider segment for us or vertical for us, there were a few quarters where their order growth was in the 100% year-over-year sort of range. So the fact that they’re going to take some time to go and to consume that inventory to deploy it is as expected. And this is also happening within the SP space as well. The other thing that I can say is, if you look at these businesses over the years, there have been ebbs and flows. They have always been lumpy. And I do not believe that there’s anything structural that is happening right now in either the SP space or the cloud space that would suggest that this is the new normal.
I expect that they will both bounce back. And as Ken mentioned, orders in both of these verticals will recover next year. In terms of the timing, the time frame, it’s difficult to say. All I can tell you at this point is that it’s going to take a few more quarters of digestion before we start to see a meaningful recovery, meaningful rebound, but I’m optimistic it’s going to come. And maybe the last thing I would say is the following. It’s the very fact that these verticals, Cloud and Service Providers, have traditionally been lumpy that we, several years ago, decided very strategically, very deliberately to pursue the Enterprise and to diversify our business into the Enterprise. And it is, in fact, now the Enterprise, which, for the first time in our history, represents over 50% of our revenue this past quarter is giving us the resilience that we need to weather challenges in SP and Cloud.
Those challenges are going to go away. It’s not going to be the new normal, as I mentioned. And eventually, we’ll have the benefit of a rebounding SP and Cloud business in addition to an Enterprise business that I believe will continue to perform well without all of the typical gyrations of the business that we have endured in SP and Cloud historically.
Operator: Thank you. And the next question is coming from Atif Malik from Citi. Atif, your line is live.
Atif Malik: Hi, thank you for taking my questions. I have two. The first one is for Rami. Rami, in your prepared remarks, you sounded quite instructive on the Ethernet adoption for AI clusters. You talked about front-end, back-end, infill storage. Can you talk about the timing of the Ethernet adoption? Is this something like 2025 event? Or are you seeing rail pilot or volume rollout?
Rami Rahim: Yeah. So I’m actually quite bullish about the AI cluster opportunity. As I mentioned in the last call, I think we all have to acknowledge that today, the technology of choice for connectivity between GPUs and either inference or learning clusters is InfiniBand. It’s not Ethernet. However, the momentum behind Ethernet in the industry is very strong. And so I believe it is a matter of time before Ethernet reigns as the fabric technology of choice in AI clusters. And I do believe this will present opportunities for us. In terms of timing, I think next year, 2024, maybe closer to the second half of the year and in particular, what we’re seeing, especially among the cloud majors customers for us, there are a lot of projects.
There are a lot of opportunities out there where we are being asked. There have been some good technical dialogue about how we move Ethernet to become sort of that cluster technology of choice for either learning or inference for their solutions. And in fact, even large enterprises, we’re seeing that more and more large enterprises, financial services, insurance companies, even healthcare, they’re pursuing these private clusters. All up, I’m optimistic about the opportunity. I like how we stack up from a technology standpoint with the combination of our customer and merchant silicon, our Junos operating system feature that we developed specifically for the Ethernet cluster solution and Apstra, which provides the automation and the visibility into it, Ethernet AI cluster solution, that gives me good optimism for capturing our fair share, if not more, in this market opportunity.
Atif Malik: Great. And then, Ken, when you guys talk about your Enterprise revenues to grow next year, is that growth mostly attributable to the market share? Or are you seeing the overall market also grow for Enterprise?
