SUSE takes aim at Red Hat in Linux, Kubernetes play

As a major supplier of enterprise Linux software, SUSE is arguably the only other company besides Red Hat that has the resources to drive open source software development.

It recently announced that it would invest over $10m to develop, fork and maintain a Red Hat Enterprise Linux (RHEL)-compatible distribution available to all without restrictions. The move follows Red Hat’s decision to limit access to RHEL code to Red Hat customers in a bid to prevent rebuilders from profiting from RHEL code without adding value to the software.

However, SUSE’s newly minted CEO, Dirk-Peter van Leeuwen, a former Red Hat veteran who oversaw Red Hat’s Asia-Pacific (APAC) business, believes open source companies should not compete on source code or intellectual property, or anything associated with it. “There’s no way of explaining that in open source,” he told Computer Weekly during a recent interview in Singapore.

During the interview, van Leeuwen, along with his chief technology officer Thomas Di Giacomo and regional general manager for APAC Josep Garcia, discussed SUSE’s recent moves and how the company plans to tap the Kubernetes groundswell through Rancher and its long-standing partnership with SAP to grow its business in the region and elsewhere.

What are your plans for SUSE and what’s your strategy to grow the business in the APAC region?

Van Leeuwen: It’s exciting for SUSE because we have a lot of great products and we are in a unique space, mainly around Rancher for Kubernetes management. We’ve also seen huge adoption of K3s, our lightweight version of Kubernetes that was built from the ground up for container management at the edge. We also have a cross-container management tool to help our customers manage a diverse ecosystem.

With SUSE, it's really about managing what the customer uses across platforms, not just SUSE workloads. And that’s what customers are looking for. We do that for Kubernetes, as well as for Linux, where they can manage any type of Linux including the popular and less popular ones.

Those two are important themes to understand because customers want choices. They don’t use one system alone, but they still need to manage a heterogenous environment. Plus, we have very deep knowledge about managing open source projects and products, working with the upstream.

So aside from Rancher, which has huge adoption, we also have NeuVector for container security, which does deep packet inspection to check what’s going on between containers and to detect things like confidential information going from one app to another that you may not be able to catch.

These products are important for our future and you’re going to see us double down on them as we go to market. We also have relationships with the open source community and large partners. With Intel, we are working on confidential computing, which we are implementing now for the next chip it will be releasing. This will enable you to run workloads in an encrypted state, so if you’re in a regulated industry, you don’t have to worry about having your systems run outside of your premises or country, because no one can see what’s running inside a container or processor.

We’re also investing heavily in high-performance computing for AI [artificial intelligence] and we’re bringing AI into Rancher so you can use generative AI to configure your container environment. These are very strategic capabilities that will help us to reposition ourselves in the market. Obviously, our branding needs to come along with it because many people are not always aware that Rancher is part of SUSE, so we need to work on that as well.

“Rancher is more adopted at the edge than anybody by a factor of 40, so it's an industry standard for Kubernetes at the edge. It’s also the only open source, multi-cloud and multi-cluster management tool for Kubernetes as well. The competition is trying to catch up, but they are nowhere near us”

Historically, we’re known for our Linux portfolio, which has been driving our strong relationship with SAP. Today. we’re the preferred platform for SAP workloads and that gets us into many large enterprises. Aside from that, we want to be that alternative choice for enterprise customers because people want choice. They don’t just want one Linux vendor and they don’t want to be locked in as you could with proprietary solutions. You cannot be locked in with an open source solution.

Providing choice was not something we came up with lightly. It was driven by customer requests – customers told us they needed an alternative. Initially, there were a few commercial requirements from customers but with the announcement about RHEL, we realised that we also needed to offer a community solution because this is not just a commercial issue; it’s also a community issue.

We’re all open source people and we’re really passionate about what we do. We believe that if you compete in an open source world, you don’t compete on source code or intellectual property, or anything associated with it. There’s no way of explaining that in open source. You can only compete on added value, which is your servers, certifications, support and all the other things that are important for customers to run enterprise-grade software. It’s not about the code, copyright or anything related to that. We feel very strongly about it and that’s why we said we’ll have to fork RHEL because otherwise there’s no choice.

With regards to your plans to fork RHEL, do you have more details since the time of the announcement and what is the plan to achieve that vision?

Van Leeuwen: The whole plan is to engage the community. If it wasn’t for the community, we’d already have a product. We have a programme called Liberty Linux, which allows RHEL customers to switch to SUSE and get their support from us for other Linux variants as well. But we wanted to go beyond that because we felt that this was not just about switching a commercial customer. This is about the core of what open source is, and making the code open again. We can take a leadership role in the community, but it’s also up to several foundations in the community to work with us and agree on how we do it.

Di Giacomo: Gregory Kurtzer, CEO of CIQ and founder of Rocky Linux and CentOS, is already working with us. We will work with all RHEL-compatible distributions, but our goal is to do that as a foundation and a community, because it’s not about any one company and that’s not how open source works. The idea is to maintain compatibility at the binary level because the ecosystem is there. But we are not dropping SUSE Linux Enterprise, so we will have two streams, and we’ve been doing ISV [independent software vendor] certifications for many years.

Van Leeuwen: We have the unique position of being the only other company that has the capacity and know-how to run an operating system lifecycle. It’s a very natural thing for us to do. It’s not hard and it doesn’t take a whole lot of additional resources. We have all the people in the community, and we are very well recognised in the open source community. Relative to our size, we are also the biggest contributor.

Now, if you look at the commercial side, customers also want choice and don’t mind paying for support. But they’re looking for alternatives. The example I always use is that when you want to switch telcos, you can keep your number but change providers. We’re offering RHEL customers exactly that – you keep your software and get your service from another provider, and that’s how we will compete.

We’ve been in this industry long enough to see these sorts of kerfuffle emerge every now and then, but it always goes back to the root of the issue, which is about balancing commercial interests with what open source is about. But do enterprise customers really care about whether the source code is open or not? What are your thoughts on managing the balance between commercial interests and staying true to who you are as an open source company?

Van Leeuwen: I think enterprise customers do care and they may not realise they care. The reason they moved to open source in the first place was to access the innovation and meritocracy associated with open source development. You may not realise it if you’re just following the lead of others, but that has driven the success of open source for enterprise customers.

Their problem has always been how do I cover the things that the community cannot cover for me? How do I cover an SLA [service level agreement]? How do I cover the long-term stability of my product and get all the updates I need so that I don’t have to reinstall software every three months because things move a lot faster in the community.

So, all the things that an enterprise open source company offers as a service are super valuable to enterprise customers. And then at some point, they scale out and they see that they have a huge number of installed systems from maybe one vendor, and then they get the traditional price lock-in because they have no alternatives.

“We believe that if you compete in an open source world, you don't compete on source code or intellectual property, or anything associated with it. There's no way of explaining that in open source. You can only compete on added value, which is your servers, certifications, support and all the other things that are important for customers to run enterprise-grade software”

And every customer wants an alternative. They don’t want to be dependent on one vendor. It’s a simple economic effect of purchasing. And there really isn’t a good alternative to RHEL, so we feel we have a great opportunity to step up there. To the extent that customers don’t care, they don’t need to care, because they will have a choice of two suppliers.

Di Giacomo: Some of them do care as well, because with the lock-in, if they don’t have the source code, or if the source code is not open, they cannot add their own software or add value to it with their partners. And the community cares and needs the source code.

You have to understand that more than 90% of RHEL was not developed by Red Hat. So, if Red Hat loses the community, they don’t have a product to sell any more and they are not relevant any more. It’s very short-sighted, even from a business standpoint, because by losing the community, you lose your products.

I think this move by Red Hat has put Linux under the spotlight once again. For some time, Red Hat has been very focused on OpenShift to the extent of overshadowing the work it has been doing with RHEL. Do you see operating systems becoming less relevant now that more people are using containers? Will it be commoditised at some point when it doesn’t really matter whether it’s RHEL or CentOS?

Van Leeuwen: Commoditisation is happening for sure, and that’s great because it means there’s mass adoption. But it doesn’t mean the innovation stops. If you run something in a container, you still need to know what’s in that container. If you do something at the edge, you still need some form of micro Linux to run your applications. That needs to evolve with all the technology that has been developed on the hardware side. So even though it’s commodity, just like chipsets, everything else doesn’t stop evolving. It continues to be very relevant, especially the open source component, which is crucial to the evolution of this technology.

How is SUSE driving commercial adoption of Rancher?

Di Giacomo: From a user standpoint, Rancher is more adopted at the edge than anybody, by a factor of 40, so it’s an industry standard for Kubernetes at the edge. It’s also the only open source, multicloud and multicluster management tool for Kubernetes as well. The competition is trying to catch up, but they are nowhere near us.

As for commercial adoption by companies, especially those in APAC, it’s still early stages in some markets. But we are well-positioned to help those companies to make that transformation. And the way we differentiate from the competition is that we are very open. It’s not about taking an entire stack and not being able to add modules from others. We don’t believe in any of that – the cloud-native ecosystem has to be open.

Rancher is widely used by individuals because it’s easy to install. You don’t need a PhD to install Rancher or take three weeks to install it on 20 machines like you do with other solutions for container management.

Van Leeuwen: The way we go to market is the biggest change that you’ll see in the short term, because we want to be present in different countries in the region and we want to make sure we have strong executive relationships with customers. Heavyweights like Josep will build those relationships and gain insights into what our customers are doing.

We used to focus more on products when we sell, but now we’re going to focus more on solutions. That will, without any doubt, drive high adoption of our entire portfolio. We’re in the process of rebuilding things to some extent. We have great people, but there have been some previous reorganisations that fragmented things a little. I’m bringing them all back together so that we can have a strong go-to-market momentum.

Garcia: In APAC, we are unifying our salesforce to make sure everybody is able to sell solutions that integrate our technology. We’re going to go deeper on building relationships with our customers and getting new customers. We’re going to grow our coverage and focus on building partnerships, particularly with systems integrators that have been using Rancher but are not working with us to bring Rancher solutions to their customers.

It’s going to take us a while and we are just getting started. There’s a lot of interest in having an alternative player that can offer the freedom of choice and allow them to work on many different platforms. One of the things we hear regularly is that other container solutions are not flexible. For example, OpenShift can only manage OpenShift containers. A lot of partners are saying that they want to build solutions using Amazon and Microsoft containers. But there’s going to be a need for a centralised management console that can manage different container environments at scale. And they see that we can provide that. So, we have a very strong value proposition not only to customers, but also to partners that are building containerised applications.

The partnership between SUSE and SAP is well known, but what is the impact of efforts by SAP to move more customers to S/4Hana on cloud?

Di Giacomo: When SAP workloads move from on-premise to S/4Hana in cloud, more than 90% move to SaaS [software as a service]. It’s a growth strip for us – we are number one with on-premise Hana workloads, so it’s net-new business. The more S/4Hana cloud migrations there are, the better it is. But as an ecosystem partner, we are also close to the hyperscalers – we have integrations with Microsoft, Amazon, Google and the cloud providers in China for SAP workloads.

A lot of SAP modernisation projects are complicated and being driven by systems integrators such as Tata Consultancy Services, Cognizant and Accenture, which work with us on those projects. The future of SAP in the cloud lies in Hana, which was developed on SUSE Linux. SUSE contributed to the Linux kernel features for Hana to run properly on Linux, so we are very happy about that.

Garcia: Many of our partners, like those in China and Japan, say there’s still a large number of customers that are running SAP on Windows and SQL Server. As those customers migrate to S/4Hana, whether it’s on-premise, on SAP Cloud or Microsoft, they are going to migrate to Hana on Linux. In APAC, many Hana deployments are happening on SUSE, so it’s definitely a significant upside.

Longer term, I think the opportunity in APAC is the container market with the shift to microservices by telcos, manufacturers and financial services firms. We are maybe – with the exception of Australia – one or two years behind what’s been happening in the US, and that’s going to be a significant growth driver for us.