Episode 413: AI and Data Modernization

Justin Parisi (0:00)

Data modernization is no longer just about where data lives. It is about making data accessible, protected, governed, and ready for the next wave of innovation, especially as organizations look to bring AI from experimentation into production. This week, Ahmed Al-Nabhani joins us to talk about AI, data modernization, and how NetApp helps solve some of your toughest data challenges. Welcome to the Tech on Tap Podcast. I love NetApp because it's so funny.

I'm here in the basement of my house and I have Ahmed Al-Nabhani with me. We're here to talk about AI stuff, like AI data pipelines, AI modernization, and how it really integrates in with the NetApp story. So Ahmed, what do you do here at NetApp and how do we reach you?

Ahmed Al-Nabhani (0:56)

Thank you so much, Justin, for having me on this great podcast. So in order to talk more about data modernization and data pipeline, I'm working as a Senior Solutions Engineer and I'm building the UAE Enterprise Center. And you can reach me out on LinkedIn or through you. I'm very glad to be in such a great team, GCC, at supporting in UAE. As a technology advisor for our customers, we are a specialist when it comes to Internet platform and pipeline. And now we are mapping the link into NetApp.

Justin Parisi (1:26)

All right, excellent. We'll include some of those links in the blog post and the post that go along with this. But let's just kind of kick things off by talking about data modernization. This term gets thrown around a lot.

So when you talk about data modernization, what do you mean by it?

Ahmed Al-Nabhani (1:42)

Yeah, actually today's conversation is all about data modernization. It's not just refreshing the storage. It's about modernization, how data is being discovered and governed, moved and shipped from way to another, and has been consumed by the enterprise. Especially when we are talking about enterprise, especially nowadays, we are talking about hybrid environments so AI can actually scale. Most of AI programs and initiatives we are seeing that stall because of data access, governments cleaning the data, lineages and operational complexity, it becomes a bottleneck, not only the GPUs.

When we are talking more about data modernization, we are talking about how we can upgrade data systems, how we can integrate it with the current infrastructure in a practice, and how data can be accessed everywhere, and how we can use it, how we can secure it and make it more valuable, and we can use it as a product. We can see that in the cloud, we can see it in the hybrid cloud architecture, and with a better governance. It's not just a migration project, it's a new operational model. So every organization needs to start how to integrate with their operation model, how we can change it, and how we can unify multiple break signals into one standardized governed automated memory. So in advance, we can say that the data modernization nowadays is a data foundation, a strategy for early modernization. The next question we can ask when the organization comes into enterprise, how do you feed your analytics in the pipeline? How you are living without AI strategy?

What is the AI affecting your data management workflow, and how you are keeping your data from multiple silos into one unified data platform? So this is what we are talking about, the data modernization and how it is affecting the enterprise.

Justin Parisi (3:35)

Yeah, there's a lot to it, right? It's not just taking some data and chewing through it and making something out of it. There's a lot that goes into it. You don't want to grab the wrong data.

You don't want to put things into your AI inferencing that maybe doesn't make sense because you'll start to get hallucinations. But then you have to start worrying about the governance, the regulations. So when it comes to the Middle East, what sort of regulations and compliance issues are you running into when we're talking about data modernization and governance?

Ahmed Al-Nabhani (4:04)

Well, we have most of the aspects actually in GCC when it comes into mapping, how we can have applying practical AI data management strategy.

We can start with the AI ready or AI fair strategy. It comes into an outcome for data pipeline. We can have trusted data access. We can have a policy. We can have audit which comes into a repeatable data pipeline from dev to production. This is really aligns with the intelligent data infrastructure strategy that currently the enterprises are going.

We can talk more about the hybrid by design. We have seen multiple enterprises that are not putting their data in multiple silos or on-prem. They are having multiple infrastructure or data centers that is on-prem or to the cloud in a different GCC or the public hyperscaler regions. We have seen a lot of initiatives that come into GCC for AI commonly need for the hybrid cloud community. We can see data stays for sovereignty inside the country. We can see the compute bears that comes into services that need to be accessed by external users, not only to the cloud. And we can position that in our reference architecture story, not only to be compromised by others. Governments for GCC is not only a security, it's an add-on for the data sovereignty. We are emphasizing the enterprise in order to have a consistency governance and guardrails for these kind of security contexts for the mobile. We have the right architecture order to have these kind of workflows to run on ONTAP, and how we can build our AI pipeline scale and lead into this aggregated story, performance and capacity, and how we can have our data modernization and what comes into GCC enabling by AI safety, and when it comes into skin, and how we can unify multiple data silos into one data plane, automated and ready for be delivered for customers. We are talking about data modernization, how it is reshaping the GCC region from modernization prediction, having unstructured pipeline, through AI moves from pilot to production.