Cloud Native

Cloud Foundry vs Kubernetes: App Platform Trade-offs | Julian Fischer, anynines | TFiR

0

For organizations running hundreds of applications, platform choice is a financial decision as much as a technical one. Kubernetes is flexible but ships without a service marketplace, without centralized data services, and without a unified operational model, forcing each team to solve problems Cloud Foundry solved years ago. The overhead-per-app gap is real, measurable, and rarely discussed honestly.

In this interview on TFiR, Julian Fischer, CEO at anynines, breaks down where Cloud Foundry still outperforms Kubernetes for large 12-factor application portfolios, where the trade-offs shift, and what the current state of the Cloud Foundry ecosystem means for teams making platform decisions today.

Guest: Julian Fischer, CEO at anynines
Show: Cloud: Evolution
Watch the discussion on YouTube

Here is what every platform engineer and application architect needs to know.

Technical Deep Dive

Q: Is Cloud Foundry vs Kubernetes still a relevant debate for enterprise platform teams?

Julian Fischer, CEO of anynines, argues the debate is not settled and the answer depends on application portfolio size and operational model. Fischer identifies a clear clustering: some organizations have lost confidence in Cloud Foundry’s long-term sustainability and are moving away, while others recognize it as the most operationally efficient platform for running large numbers of 12-factor-compliant applications. The choice is not ideological; it is a calculation of overhead per app.

“Cloud Foundry to this day is a turnkey solution for building an application developer platform. You don’t have the clutch problem.”

Julian Fischer, CEO, anynines

Q: What does Cloud Foundry provide out of the box that Kubernetes does not?

Fischer identifies the built-in marketplace, Open Service Broker API, and managed data services as the core differentiators. With Cloud Foundry, a team gets a centrally provided, fully operational platform without training each team independently on how to run it. Fischer describes deploying anynines’ commercial Cloud Foundry environment on any cloud with anynines data services as a configuration that simply runs with high uptime and no recurring operational emergencies.

“If you have a Cloud Foundry, you have a marketplace, you have the Open Service Broker API, you have the nice data services, you have a turnkey solution.”

Julian Fischer, CEO, anynines

Q: Is Kubernetes cheaper than Cloud Foundry for running large application portfolios?

Fischer is direct: Kubernetes is not cheaper and not more efficient for this workload profile. The flexibility Kubernetes provides comes at the cost of unsolved operational problems that Cloud Foundry handles centrally. Fischer frames it as a trade-off: Kubernetes offers adaptability; Cloud Foundry offers financial economy at scale. Teams must calculate the real overhead per application before assuming Kubernetes is the default cost-effective choice.

“It’s not cheaper, it’s not more efficient. It leaves organizations with more unsolved problems than Cloud Foundry does, but it provides more flexibility and adaptability.”

Julian Fischer, CEO, anynines

Q: How stable is the Cloud Foundry ecosystem today given uncertainty about the foundation?

Fischer acknowledges uncertainty about the Cloud Foundry Foundation and overall maintenance trajectory but points to strong commercial backing as a stabilizing force. Broadcom has doubled down on Cloud Foundry, SAP remains a strong participant, and Fischer notes that other companies are joining the community and that events are growing again. Fischer positions Cloud Foundry as a stable alternative to Kubernetes specifically for 12-factor applications requiring high financial economy of operation.

“Broadcom has doubled down on Cloud Foundry, SAP is still strong with Cloud Foundry, and other companies are joining the community. The events are growing again.”

Julian Fischer, CEO, anynines

Q: When should an organization choose Kubernetes over Cloud Foundry?

Fischer is candid that Kubernetes is the larger, more dynamic, and more commercially appealing market for building products today. Organizations with strong Kubernetes demand internally, or those that can absorb the operational overhead, have a valid case. The key variable is whether the organization needs the flexibility and adaptability Kubernetes provides enough to accept the higher operational cost and the burden of solving problems Cloud Foundry handles automatically.

“Obviously Kubernetes is the bigger, more dynamic, more appealing market to produce products today, so that’s also crystal clear.”

Julian Fischer, CEO, anynines

Resources & Documentation

  • anynines, commercial Cloud Foundry environments and managed data services deployable on any cloud
  • Cloud Foundry Foundation, open source platform with marketplace, Open Service Broker API, and managed data services
  • Open Service Broker API, open standard for integrating managed services into Cloud Foundry and Kubernetes platforms

***

👇 Click to Read Full Raw Transcript

Swapnil Bhartiya: I mean you have been in Cloud Foundry world for a long time. How, I mean we have talked about this a lot in past and I want to talk about one more time, maybe last time, how are the Cloud Foundry users evolving and moving to Kubernetes? Because when I look at Anynines, when I go to the website, all the anynines services, I mean not the service but all the offerings that you have, they’re more geared towards Kubernetes workloads. So talk about that evolution that you’re seeing there and how Anynines is best positioned to help those customers.

Julian Fischer: That’s a good question. And I really have to sort my thoughts on this one because it’s quite controversial. What people do, it’s not controversial, but there’s a clustering of what people do. Some people see that Cloud Foundry is in the mice and move away from it. They just lost their belief in that this is a technology that is sustainable. Well, and for some organization that makes sense because they can get away with not having Cloud Foundry. But if you’re looking at organizations with a lot of application developers and they have 12-factor compliant apps, and I’ve been saying that for years, Cloud Foundry to this day is a turnkey solution for building an application developer platform. You don’t have the clutch problem. If you have a Cloud Foundry, you have a marketplace, you have the open service broker API, you have the nice data services, you have a turnkey solution. We basically take commercial Cloud Foundry environments. Put it on any cloud, only with anynines data services and it runs and runs and runs uptime. Great. There’s no outage here and outage there because you have to maintain this and maintain that. You don’t have to train each team on how to run Cloud Foundry. It’s centrally provided and it works at scale. So if you take that operational efficiency and you calculate how much overhead you have per app, there’s no better solution for running a large number of apps than Cloud Foundry. To this day, I believe still believe this is going to change, but interestingly it did not happen yet. So where I see. And then consequently we have customers looking at this fact very clearly. Of course there’s uncertainty about what happens to the Cloud Foundry foundation and the overall maintenance of Cloud Foundry. But so far we have good and strong companies behind it. I mean, Broadcom has doubled down on Cloud Foundry, SAP is still strong with Cloud Foundry and other companies are joining the community. The events are growing again. So we basically see the Cloud Foundry stack as a stable alternative to Kubernetes where for 12-factor applications. A lot of them need to be operated with a high economy, financial economy. Now, organizations, all of them, have Kubernetes demand internally. And it’s absolutely okay to look at application developers and tell them Kubernetes is the way to go. It is possible, but it’s not cheaper, it’s not more efficient. It leaves organizations with more unsolved problems than Cloud Foundry does, but it provides more flexibility and adaptability. So it’s a trade off. If you ask me about the market. Well, obviously Kubernetes is the bigger, more dynamic, more appealing market to produce products today, so that’s also crystal clear.

Why AI Inference Fails at Scale and How Distributed Infrastructure Fixes It | Jon Alexander, Akamai | TFiR

Previous article

Right-Sizing AI Models for Enterprise Cost | Tony Thomas, Zoho Corporation | TFiR

Next article