What is our primary use case?
We're the largest financial institution in Africa, and we use various operating systems and technologies to achieve typical financial service goals. In the past, we were an ION-centric shop. However, in the past decade, we've been increasingly leveraging Linux's agility compared to traditional Unix operating systems.
Generally, we deploy by cloud, but we use RHEL on-premise in our data centers and prefer SaaS for infrastructure as a service. Our primary cloud providers are AWS and Azure, and we also use smaller third parties for niche environments.
RHEL is spread across virtually all elements of the institution, including headquarters and various locations on multiple continents. In my environment, it is part of a global trading settlement system.
The rollout for this particular solution was probably about 250 users of the application running on the initial RHEL. We're a global bank, so the user base is much larger worldwide. Users include business and feature analysts, engineers, and project managers. Our infrastructure engineers were the ones pushing for a switch to RHEL, followed immediately by application engineers.
How has it helped my organization?
RHEL enabled us to move away from reliance on ION. We're free to choose the best-of-breed solution at any given time while keeping the cloud-agnostic infrastructure at the center of our deployments.
Our operational expenditures decreased, and RHEL made our teams much more flexible. With RHEL, we can have multiple copies of an OS without making annual plans to license and acquire.
The benefits were instant from my team's perspective. For example, we were immediately more flexible and able to scale rapidly. However, if you're looking at it from an executive point of view, the time to value depends mainly on the product and the scale of the endeavor. It might take a few years to reap a return. Ultimately, you will see the financial benefit, but that's somewhat difficult to quantify in the short term.
I don't think that it's enabled us to centralize development, but it has perhaps increased the breadth of development possible on our applications. In that sense, more development can be centralized on the operating system, but that's more of a byproduct.
We outsource cyber security to other teams, so I can't comment in-depth on RHEL's security features, but I can say it enabled us to understand our security posture more efficiently. This wasn't always possible using an AIX or Solaris in a more centralized fashion. The feature set is maybe not as important as having a single pane of glass and a single configuration to apply across our systems and infrastructure.
RHEL made life a lot easier in terms of compliance because you can more accurately gauge yourself against industry benchmarks with the tools provided and identify your shortcomings. You can interrogate what you've done through research from multiple parties rather than just a single source of truth, which may not be true.
What is most valuable?
You can compile and run applications on any operating system, but RHEL's advantage is flexibility. It is more supported and supportable in the enterprise sense than Ubuntu or perhaps a smaller distro, but it's also flexible enough to easily transport from platform to platform: ISA to ISA, production to development, and vice versa. That led me to embrace the switch to RHEL from other operating system variants.
RHEL offers more portability than any other OS flavor apart from perhaps Ubuntu Linux. As a large bank, we run on IBM's architecture. We run Power, Spark, and Oracle x86 across multiple environments. It lets us choose the right environment for the application, which is essential from an operational efficiency perspective. These days, we're all trying to cut heavy infrastructure and move to lightweight agile infrastructure. There isn't a better option in the production world than Red Hat.
What needs improvement?
There needs to be a broader understanding of the RHEL suite's nuances like how the versioning works and implementing it on various kinds of infrastructure in use across the development landscape. There needs to be more training and education. It's difficult when you have a roadmap to deal with, but it is possible.
Large application vendors may not have certified RHEL, or they have certified an older version. Most of the large application vendors are unfamiliar with the versioning that RHEL introduced, which I strongly support. They will support a given sub-version up to a point, not realizing that the sub-versions are essentially additive.
This can be a real frustration when you try to deploy modern infrastructure. It allows tremendous flexibility because we can try things out across the cloud, virtual, and physical, but that's not always where the issue is. It's a matter of educating the engineers and developers on our side or enterprise vendors on the other.
The licensing could also be simplified. While it makes sense from a theoretical perspective, it's a challenge to explain to the procurement team. Those with some technical expertise can understand how our licensing model works. However, it's still tricky because Red Hat is so different from traditional operating systems. It's another barrier when I'm trying to deploy it in an enterprise environment.
In terms of feature requests, I would point out that our company tends not to operate on the bleeding edge for obvious reasons. We look at what has already been released to define our roadmaps. There's nothing in particular that I would say needs to be included. However, I would like to see Arm playing a more prominent role in the cloud infrastructure and enterprise physical data center spaces. Red Hat supports this, but I haven't seen a clear roadmap for how that support should evolve within the Red Hat operating system environment.
For how long have I used the solution?
I have used Red Hat Enterprise Linux for more than 10 years.
What do I think about the stability of the solution?
RHEL's stability is good.
What do I think about the scalability of the solution?
RHEL is highly scalable and we plan to increase usage.
How are customer service and support?
I wouldn't rate Red Hat support as less than eight out of ten because I can't think of anything negative to say. I can't think of a time when I haven't been able to get it. Also, because RHEL is global and Linux is open-source, you can typically get the support that you need through research forums and the knowledge base. It's seldom necessary to involve third-tier support within RHEL.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
We still use other operating systems. We've used just about every solution you could name in conjunction with RHEL. We also deploy Ubuntu. In some cases, our application vendor requires us to stick with a given solution. Sometimes it's AIX or Solaris, but mostly we can override that and move to RHEL. Red Hat is now standard for most future enterprise deployments, and we run RHEL on mainframes too, but in a very limited fashion.
How was the initial setup?
The setup was complicated only because the applications we were trying to run were not certified to run on RHEL. It was version 6.8, so we worked with major global vendors to add the certification for the versions we were trying to run. That was the complexity. The application always worked beautifully, and the performance was excellent. It wasn't a question of getting the development to work; obtaining an issue of getting certification for the platform, which is required for any financial institution.
From a development perspective, we proved the concept and ran a mirror of production and development to demonstrate the improvements in OpEx and performance. Getting it up and running in parallel was the key to getting it all to work correctly, and it was instrumental in convincing any dissenting voices of the value.
The deployment took less than three months, but the certification took nine.
The team supporting the first application numbered around 50, and the small group involved in the initial switch had about eight people.
The entire application is run exclusively on RHEL, so the whole operation team is probably around 40 or 50 people. It's worth adding that our overall group runs about 20,000 servers, so it's challenging to say overall what the RHEL footprint is.
After deployment, RHEL requires maintenance to keep the solution up to date. Security requirements tend to be more prohibitive or less encouraging of change. It's a question of changing mindsets and explaining that something doesn't have to be legacy-tested to update. The security benefits of updating are more critical than testing to ensure the update hasn't introduced more flaws.
What was our ROI?
I don't have the data, but we have significantly reduced operational expenditures since switching to RHEL. It was a reduction of more than 10 percent.
What's my experience with pricing, setup cost, and licensing?
The licensing is tricky to understand. Enterprises want to be beyond reproach when it comes to licensing. We would rather over-license than under-license. However, that can be complicated with a high-performance development team who may need multiple operating system instances or want to experiment with spinning up many machines to see if something works or sticks.
We don't necessarily need support for those. Our procurement team is confused if we need a license for an instance that was only up for 15 minutes on Thursday. We need to make sure that we always have sufficient licenses. That misunderstanding of how cloud development works can sometimes slow down development. It inhibits the growth and success of Red Hat Enterprise Linux globally. So more education around that would be beneficial or at least will provide more clarity.
RHEL's total cost of ownership is difficult to quantify, but it's almost irrelevant. In cases where you don't care, you can always use an open-source OS. In other cases, you need the support and certification that comes with something like RHEL. I do not believe RHEL has any competitors in our use case.
What other advice do I have?
I rate Red Hat Enterprise Linux nine out of ten. My advice to prospective users is to try RHEL out and see if your application works. In the long run, the benefits will outweigh the time and effort spent migrating. The important thing is to ensure you run programs in parallel so you can accurately evaluate the benefits and make a case for switching.
Which deployment model are you using for this solution?
Hybrid Cloud
Disclosure: My company does not have a business relationship with this vendor other than being a customer.
