KubeCon + CloudNativeCon EU 2026 | Francesco Giannoccaro, UK Health Security Agency

Clips
More from KubeCon + CloudNativeCon EU 2026

Francesco Giannoccaro

Head of High Performance Computing

UK Health Security Agency

play_circle_outline Trusted AI and Accelerated Cloud Computing for UKHSA Scientists: OpenShift, Signed Containers, and Baselines

play_circle_outline Simplifying Complexity: Auto-Scaling, Pipeline Automation and Hybrid Multi-Cloud Portability with Data-in-Use Security

play_circle_outline Kubernetes Scheduling for Scalable Genomics: Balancing Data-Intensive Storage and Compute-Intensive Modeling

Info
Transcript

Francesco Giannoccaro, UK Health Security Agency

Francesco Giannoccaro

Head of High Performance Computing UK Health Security Agency

In this interview from KubeCon + CloudNativeCon EU 2026, Francesco Giannoccaro, head of high-performance computing at the UK Health Security Agency, joins theCUBE Research's Rob Strechay and Rebecca Knight to discuss how cloud-native platforms are enabling public health scientists to run genomic-scale workloads without deep software engineering expertise. Giannoccaro traces the agency's journey from an on-premises OpenShift deployment in 2017 through a pandemic-driven migration to Azure in 2020 and into the current era of GPU-accelerated machine learning and ... Read more

explore Keep Exploring

What is the scale of the responsibilities you oversee at the UK Health Security Agency? add

How revolutionary was it to empower public health scientists to run and iterate real-time models for issues like pandemics, and what infrastructure, operational, and security challenges did that entail? add

What challenges do scientists face when analyzing large genomics datasets, and how can Kubernetes help overcome those challenges compared with traditional high-performance computing? add

bolt Powered by CUBE AI

Francesco Giannoccaro, UK Health Security Agency

search

Rebecca Knight

>> Good afternoon everyone, and welcome back to theCUBE's live coverage of KubeCon EU, CloudNativeCon EU. I am your host, Rebecca Knight. I'm sitting alongside Rob Strechay. Rob, we're in the final stretch of the day.

Rob Strechay

>> Final stretch. I think it's been wonderful. I think the energy that is here, the people that we've been talking to and the sharing in the community has been fantastic.

Rebecca Knight

>> Indeed. Well, with that, I would like to welcome our next guest to the show. He is Francesco Giannoccaro, Head of High Performance Computing at the UK Health Security Agency. Welcome, Francesco.

Francesco Giannoccaro

>> Thank you. Great to be here.

Rebecca Knight

>> Yeah. The UK Health Security Agency, a bit of a mouthful, is essentially the organization that the country calls when something goes wrong on a population level. Before we even get into the technology, can you talk about the scale of what you're responsible for?

Francesco Giannoccaro

>> Yes. The UK HSA is the UK expert body responsible for protecting community from infectious disease and public health emergency. We do that relying heavily on science, data, and operational capability. My role is in high performance computing and scientific computing. I focus on enabling our scientist, analyst, and public health team to run complex workload in a secure way. The scale cover both the four nations part of UK, but obviously the organization has engagement and collaboration with all the other body around the world.

Rob Strechay

>> I mean, I got to see you talk yesterday, and you did a fantastic job of breaking down just the important work that the scientists that are working on top of the infrastructure you support do. You had this one slide with the different timelines and talk to some of that timeline because I thought that was such a telling thing about how the technology enables and the different eras that you went through.

Francesco Giannoccaro

>> Yeah, so the journey that we have been going through with in general, open source technology and specifically with the OpenShift technology has been following some of the more complex service that the organization deliver, things like pathogen genomics, modeling. These are complex type of application and work the organization develop.

Rob Strechay

>> And which is all AI. Before there was AI, there was all the genomics and everything else, and you talked about what if it had been patented and things like that.

Francesco Giannoccaro

>> Exactly. The roadmap has been a little bit following how in the last five years pandemic and large outbreak has brought pathogen genomics at the very center of life science. We started in 2017 providing our scientists that don't have a software engineering background a platform where they were able to experiment prototyping their application and workload in a very efficient way, trying to simplifying the complexity of cloud native application while not finding compromising around security or scientific rigor. For us, in 2017, OpenShift was not mature as was today, as it today. Obviously Kubernetes in general has evolved and that complexity has become more manageable. In 2020, when pandemic happened, we moved the application that we were hosting on OpenShift on premise in Azure to allow more resilience and scalability. And a few year after, we started increasing the workload focus on machine learning and AI enabled analysis. We have integrated significantly hardware specific for that type of workload from NVIDIA, for instance. OpenShift AI has brought the same level of resilience and consistency that we have been experiencing in traditional application also in that space. In one way, we've been relying on a library of container image that are constantly analyzed for security patch, and that give us that comfort, that we use baseline that is already coming from a trusted source. And equally on the space of artificial intelligence, the aim is to give our scientists a way to assess, evaluate those open models that are constantly evolving with that trust that are already went through an assessment of trusted company like OpenShift RedHat or NVIDIA.

Rebecca Knight

>> How revolutionary was it for these public health scientists to be empowered in this way that you're describing by allowing them to, as you say, see that these models as they're evolving in real time and be working on really important issues like pandemics?

Francesco Giannoccaro

>> Yeah. I think the evolution has been... So the challenge I would say of that journey has been providing, again, our scientists that their background is epidemiologists, biologists, bioinformaticians, an environment where the complexity, again, related to auto-scaling, for instance, or to have pipeline automatically self-deploy patch container image, that is the aspect that was for us the most challenging one. Again, simplifying complexity and innovation. I think the community around Kubernetes and OpenShift has been, in the last few years, looked more and more closely to the scientific community than the traditional software engineering community. I feel that way, at least that is how I perceived.

Rob Strechay

>> Yeah. I mean, I think it is, and I think you're right. And I think a lot of people, I think there's more organizations like you and trying to make it the infrastructure transparent to that end user so that they can go and do their work, and do the hard work that they already, without having to worry about the pipelines and connecting it all, or how do I change models? Do you work with those to make sure you're almost like they're your customer, I would assume, and that's the approach that you take?

Francesco Giannoccaro

>> Yeah. And obviously, the other challenge are around giving our scientists an environment that allow them to experiment in confidence and ensuring that our information that are often including sensitive data are operated in a very confidential way. And then giving those prototype maybe the ability to scale on large environments that cloud and hyperscaler offer. That portability of hybrid multi-cloud platform that is consistent both on premise and on cloud has also been a big plus for our scientists. The focus and attention on security and confidentiality is very important for the organization because as I say, data often include patient sensitive data. In this space, I think the technology landscape has been maturing, and has had, in addition to encryption at rest and in transit that has been already available for many years, more recently we have been looking or the technology community has been looking to encrypt data news. This is one of the area that is coming and maturing currently and is one of the area that we are looking to implement and adopt.

Rebecca Knight

>> A lot of government and healthcare organizations talk about hybrid cloud, as you said, but they struggle to make it this one coherent system. What has made the difference for the UK Health Security Agency, would you say?

Francesco Giannoccaro

>> UK Health Security Agency is an organization that has been for, during pandemic, merging three existing government body, Public Health England, the Joint Biosecurity Centre and NHS Test and Trace. The latter two, were created during pandemic and were established completely on cloud where Public Health England was created over 10 years on site where we have our laboratories. That merge of those three organization gave us a quite broad landscape of different environment neutral technology because those two new agencies were created on cloud from scratch at speed during pandemic. So having a platform that standardized the way our scientists operate has been essential, and with flexibility often comes more complexity and that is the space and the area where I feel the community has been working hard in simplifying that complexity.

Rob Strechay

>> Yeah, because I think to that whole question and your answer, I think what's really neat about what you do is because there's a lot of physical stuff that actually has to happen. You have samples, you have machinery that's pumping out terabytes of data per run and things like that. And I know in the pharma space and other spaces, I mean, you could end up with a hundred terabytes in one day of data. How has just been able to... I think your job is extremely difficult because you have to deal with not only the physical IT assets, but physical other things that may not be connected very well or have gotten connected better over the last few years, but really you need to have these workflows that allow them to work where they need to. How much of you and your team is spent around, "Hey, we can help automate how they, if they go here," almost maybe not self-service, but getting them closer so it's easy for you?

Francesco Giannoccaro

>> Yeah, you're absolutely correct. Especially in genomics, the amount of data that are generated in labs is in order of several hundred of gigabyte, often terabyte in a day. The ability to analyze quickly those volume information has traditionally been very constrained into very complicated environment. Traditional high performance computing is definitely not an environment where scientists are immediately able to work on because the interface in those environments are common line only, and really have a very deep learning curve. Kubernetes has, over time, as I say at the beginning, work I think more closely with the scientific community, trying to look at those challenges, trying to simplify how it's possible to scale and distribute complex workload in an environment that often on premise are data intensive and some other time for modeling are just compute intensive. The scheduling and the ability of covering different type of requirements is something that has been making a difference in Kubernetes today.

Rebecca Knight

>> It sounds like what you're describing is really a blueprint and best practices for other organizations, but I can also imagine many people watching who run, who are public health leaders like you, nervous about moving this kind of sensitive information and data to the cloud. What is your advice, insight, or something you would want to tell them to ease their mind?

Francesco Giannoccaro

>> Sure. I think this is a space where today you have much more trust compared with a few years ago and is building those safe and secure environment is becoming relatively easily. There are large community and groups specified on scientific computing that I suggest to engage with. These type of events are, I think, a very good point to start talking with expert or with colleagues that have going through the same challenge, but I think is finding the right balance between looking up cutting edge technology and technology that are becoming mature. And that is striking that balance is basically the best spot to be in, where maturity meet innovation.

Rob Strechay

>> Yeah. No, I mean, I love that advice because I think that most people... I was at GTC last week and I'm like, and they're announcing Vera Rubin, but you can't even get those chips. And it's like you start to look at how people want the newest with OpenClaw, now NemoClaw, and now all these other claws. Claw this, claw that, right? So I love that advice. When you look at this over the next year, when we're talking again in-

Rebecca Knight

>> Barcelona.

Rob Strechay

>> Barcelona, yes.

Francesco Giannoccaro

>> Yeah.

Rob Strechay

>> Which would be awesome. I love both countries, so I'm very excited about that. What do you hope you can say next year that you can't yet say today?

Francesco Giannoccaro

>> I hope that by then we will be able to provide our scientist in the space of AI enable analysis, the same level of maturity that we have been providing over the last five years in traditional cloud native Kubernetes tag of application. I mentioned the fact that we, for instance, in UK, just say we have a library of trusted container image signed by us and curated and constantly patched. I hope that in the next year, we will be able to provide a catalog of AI models, open models that have the same level of maturity and trust where our scientists can confidently assess accuracy, perform, fine-tuning, and build the new and next coming application in the space of public health science.

Rebecca Knight

>> I love it. Inspiring.

Rob Strechay

>> I love it.

Rebecca Knight

>> Francesco, thank you so much for coming on theCUBE. A really interesting conversation.

Francesco Giannoccaro

>> Thank you for having me.

Rob Strechay

>> Yes, thank you.

Rebecca Knight

>> I'm Rebecca Knight for Rob Strechay. Stay tuned for more of theCUBE's live coverage of KubeCon, CloudNativeCon EU. You're watching theCUBE, the leader in enterprise tech news and analysis.