Karissa Breen [00:00:10]:
Welcome to K Beyond the Go. I’m coming to you from Oracle AI World. Yep. Formerly known as Oracle Cloud World, right here in Las Vegas at the Venetian. This is where big conversations around cloud and AI are actually happening. Beyond the hype on the ground, where the tech meets reality. This week I’m talking to people shaping what’s next from how governments can do more with less when public trust is on the line, to supporting modern workloads and what it actually looks like. Stay tuned for the inside scoop as KBI Media brings you all of the highlights.

Karissa Breen [00:00:43]:
Let’s get into it. Joining me now in person is Jenny Tsai-Smith, Senior Vice President Overall Database Product Management at Oracle. And today we’re discussing Oracle’s move towards supporting modern workloads and their AI play. So Jenny, thanks for joining me and welcome.

Jenny Tsai-Smith [00:01:05]:
Thank you, Carissa.

Karissa Breen [00:01:06]:
Okay, so Jenny, I’ve looked into your background. You’ve worked at Oracle a little over 32 years and you’ve seen some transformations in the industry generally speaking, but also for Oracle. So perhaps walk us through some of the key database announcements that has happened and will be happening over the course of the week and please tell us what’s on your mind.

Jenny Tsai-Smith [00:01:26]:
Well, first of all, we announced Oracle AI database 26 AI and this is probably most major release in a couple years. We have architected AI into the core of data management. This is essentially an AI native next generation database and with this new release, customers can build AI solutions very, very easily. They can also benefit from the increased productivity because the product and the cloud services built on Oracle AI Database is much simpler to operate. The other big news is we’ve also released the Oracle Autonomous AI Lakehouse which allows us to offer an open data platform for customers who want to not only use Oracle AI database, but also open table formats such as Apache Iceberg to basically benefit from the cheaper storage available through cloud object storage. But combine that data with the data that they already have in Oracle Database as well as other data sources and then on top of that we have a unified catalog delivered through AI Lakehouse so that customers can very easily combine data from various sources within a single catalog and benefit from faster access to all of that fabulous asset that they have.

Karissa Breen [00:02:49]:
And would you say given your tenure, your background, your pedigree in the space. What I’m often hearing at vendors like Oracle is that we’re in a really exciting time at the moment. There’s a lot of, in terms of velocity, a lot of things changing. What are your sort of thoughts around that? How this really going to transform the way people are leveraging databases, but also their businesses.

Jenny Tsai-Smith [00:03:07]:
Right. So AI, of course, changes everything. I think that’s what Larry Ellison says, and I believe it, and I see a lot of evidence for that. For one thing, you can now use natural language to get information out of your databases. So essentially, you’re increasing the number of users of these databases by hundreds, maybe thousand fold in terms of just that sheer number of access. So your database has to be able to perform much faster and scale much wider. So that’s one big thing. The other thing is that people not only want to query data in the structured relational databases, they also want to be able to combine that with unstructured data, such as PDF documents, images, videos, audio clips, anything that you could think of as information should be searchable and should be queryable easily across all of those different types.

Jenny Tsai-Smith [00:04:03]:
And that’s what we offer in the Oracle AI database.

Karissa Breen [00:04:06]:
And you mentioned before about unstructured data, and that’s something that I’m hearing a lot, generally speaking, in the industry. And you mentioned before, operative word like performance. Would you say, given the current climate that we’re in with businesses, try to stay relevant, they are really looking for that performance and making sure that they can get things at a faster pace, they can query all the stuff in their database and to really stay ahead of their competitors.

Jenny Tsai-Smith [00:04:29]:
Right. So I think somebody in Harvard actually said that AI is not gonna take away your jobs. It’s the people, the humans that use AI that will take away your business and possibly your jobs. And the idea is that with AI, everybody’s using this as sort of the booster, if you will, to help them, as you said, reach decisions faster, get better insights quickly, and then, of course, improve productivity. So there’s some component of cost savings as well. And by saving costs in certain things, you can invest in other areas to grow your business. So, yes, so there’s definitely, I think of it as an AI race to reach that goal that everybody’s working towards.

Karissa Breen [00:05:09]:
So one thing that’s interesting that I’ve observed is a lot of senior people that I’m interviewing on the customer side are saying, like, we need to do more with less now. So how does that sort of sit with you, Jenny, in terms of given everything you sort of just said, the way Oracle is innovating, the way they’re approaching, you know, modern workloads, et cetera, how is that going to overall help people sort of do more with less?

Jenny Tsai-Smith [00:05:30]:
So we have been building this Oracle database AI database engine for nearly four decades, and I know, this is a new release and we’ve done a lot of work to architect and add AI capabilities. But this database engine has been essentially time proven, time tested, and this is what people are going to need as a foundation in order to be able to run all of the workloads that they’re going to have to handle because of the demand for AI enabled processing and so forth.

Karissa Breen [00:06:02]:
And so just extending on that a little bit more, how do you sort of see like the hybrid multi cloud edge? Everyone’s talking about it. How do you sort of see that fitting together?

Jenny Tsai-Smith [00:06:11]:
Yeah, actually, before we jump to that, there’s another component that I want to mention which is that this database engine is built on what we call converged architecture and it saves costs because as I said earlier, people want to search not only the relational, you know, structure data, they also want to search other kinds of data types such as spatial graph analytics and as I said, documents, images and so forth. And we have a single database engine that can support all of that. So it eliminates the complexity of integrating all these different kinds of data types. It handles multiple kinds of workloads. You train people once to operate and recover and protect this one database, secure this one database engine, and they could apply that same layer learning for any kind of workloads that you want to run on the data. So you’re saving in terms of staffing costs as well, you’re leveraging existing skills, eliminating the complexity, the cost of integration, and then ultimately it gets you the results that you need faster because you’re eliminating that extra time that would have been required to write the code and build the solutions that you need, because everything is already integrated within a single platform. So that’s kind of the second part of the answer that I wanted to give with regards to how do people benefit and save costs and so forth. With Oracle AI Database, you’re talking about multi cloud.

Jenny Tsai-Smith [00:07:38]:
So of course you probably have heard that we’ve got partnerships with all the hyperscaler clouds and we continue to add capabilities and services onto these clouds. So today we announced the availability of our zero data loss recovery service on both Azure and Google Clouds. We have new partnership programs where customers can purchase these database services on these hyperscaler clouds through partners, through our partners, mutual partners. So there are a lot of ways that customers can now benefit. And what’s even better is that our latest release, Oracle AI Database 26 AI, is available on these hyperscaler clouds. So for customers who’ve already made decisions on which cloud they want to standardize on, but they want to continue to use Oracle AI database services. They don’t have to basically they don’t have to give up something to get the other, they get both. And for our customers who are developing applications on Azure and want to be able to use Oracle database, they don’t have to deal with network latency problems.

Jenny Tsai-Smith [00:08:44]:
So there are so many benefits that our customers are getting out of our multi cloud solution. And then for customers who are worried about leaving their data center, we have the cloud at customer, so we have exadata cloud at customer solution. So they could keep their data in their data centers but still benefit from the cloud operation efficiencies that they can gain.

Karissa Breen [00:09:07]:
So before we move on, I just want to stay with you on the converged architecture. So I’ve come from a banking background, one of the largest banks in Australia. Are we obviously familiar with Oracle. So how would you approach sort of a traditional player with banking records that have got, you know, very legacy sort of systems, technical debt? How do you sort of approach that in sort of transforming this? It’s easy for us to talk about it in theory, but when you’re dealing with a bank that is hundreds of years old, it’s a lot harder in doing so. So I’m keen to sort of get your thoughts on that.

Jenny Tsai-Smith [00:09:37]:
The advantage that they have if they select Oracle is that we have sort of this multi dimensional deployment strateg that’s available to them. So you know, some of these customers that we deal with are on mainframes and they want to be able to start with taking mainframe to some other platform on premises and then at some point in time move to the cloud. So they could start with on premises exadata, for example, our engineered system, we have a team of people who can help them migrate both not just the data but also the code. Because application code tends to be the sort of the long pole in terms of migrations. And so we have partners and people on the team who could help them move that data over to Oracle database on premises and then from there they could take it to the cloud and so forth. The other thing is that with AI we have customers who are actually, you know, experimenting with AI in the migration process. So what they do is they take their code, they use AI to help them understand the code. Because some of this code, as you said, is decades old and the people who wrote them may not have documented the code very well.

Jenny Tsai-Smith [00:10:48]:
Right. And they’re not around. But what we could do is use AI to parse the code and understand get an understanding what the code is trying to do. And then once that understanding is there, we could take that as sort of the specification for the application, if you will, and put it into code generators. And of course, you’ve probably heard of a lot of AI code generators. And in fact, Oracle is also developing a low code, no code AI application generator using our low code application platform called Apex. But that’s another whole big story there. But anyway, so for our customers who are on more legacy systems, we have both the technology and in terms of resources and services that will help them move to Oracle database and modernize.

Karissa Breen [00:11:37]:
So given what you’ve just said, would you say, historically speaking, customers like banks, for example, were apprehensive about doing this because like you said, someone from 40 years ago coded something, the guy doesn’t work there anymore. We can’t understand it. It’s a really old language we don’t know. Would you say that was a big risk for people to be like, we don’t want to modernize because we can’t lose this and we’re just praying that things sort of work out?

Jenny Tsai-Smith [00:11:59]:
Exactly. I think the mantra is if it’s not broken, don’t fix it. Right. And so that’s been the modus operandi, I guess, for many of these customers. But they realize that if they do that, they’re missing out on all this opportunity, especially with AI and so forth. And so for a lot of them, it’s sort of a, they have to do it on a competitive level to be able to stay ahead of their competitors. Right. Or at least keep up with their competitors.

Jenny Tsai-Smith [00:12:26]:
So. So that’s one of the reasons why they’re starting to move. And then also, of course, like I said, we have tools, we have services that will help them move.

Karissa Breen [00:12:35]:
So just on the opportunity, saying competitive, for example, would you say as well that because of this market that we’re in, it’s super aggressive. People, as in companies can’t take a long time to make these decisions in terms of, historically speaking, we’d sit around a room, we’d take 12, 18, 24 months to make a decision. But nowadays people don’t really have that sort of time on their side because of how aggressive it is to stay relevant. And we’re seeing companies come from literally zero and are overtaking these big enterprises. So executives now sort of just having to move with agility rather than doing a big risk assessment and getting everyone in the room.

Jenny Tsai-Smith [00:13:16]:
Yes, I think that is today, you can’t really avoid that. Right. I think we like to call it AI for data. Revolution is here and you not only need to survive, but you also need to figure out how to thrive. So definitely they have to take the leap, if you will. But like I said, Oracle AI database gives them that path, so they don’t have to make that big leap all at once. The other thing too is because we’ve built in all these capabilities in this single engine, they could start small, they can use a portion of it, they’ve already paid for everything, so they don’t have to later come back and say, okay, now we have to write another purchase order for this other piece of technology, because again, they already have it all in one.

Karissa Breen [00:13:56]:
So it’s a fair assumption that Oracle would have the monopoly in terms of databases. But I’m keen to maybe understand what differentiates Oracle’s database offerings versus other DB vendors. And maybe because you’ve been at Oracle for so long and you’re the database expert, is there something that is a bit more nuanced that people don’t know about what differentiates Oracle from other players?

Jenny Tsai-Smith [00:14:18]:
You know, part of it is because we’ve been around for such a long time what we’ve done, you know, decades ago that were leading edge, that are still leading edge, but people just forget about some of these things. So, for example, the Engineer system that we shipped, I want to say, almost between 10 and 15 years ago, the Engineer system, which is the Oracle Exadata database machine, we basically pioneered that. And as far as I know, nobody really caught up to us there. Well, we took hardware and software and we essentially integrated the two tightly coupled with the Oracle database engine. And so it is the most optimized platform to run Oracle database. And we continue to add capabilities. For example, in the AI space, Oracle XDATA outperforms any of the platforms that we’ve tested, where if you need to do optimization for indexing, creating indexes for the vectors in the AI database, or creating the embeddings to support the vector database. So all that is optimized.

Jenny Tsai-Smith [00:15:18]:
And we have seen so many customers today, I think over 70, maybe 80% of the customers today who are in the Fortune 100 list, they use Oracle Exadata to run their business. And probably about 60% of them are in the cloud. So I think that’s a testament. And again, sometimes I meet people and I say, Oracle Exadata. And they say, what is that? And I’m just shocked because it’s been out there and so many companies that are huge companies use it. So we probably need to do a better job of promoting some of these customers and use cases.

Karissa Breen [00:15:54]:
So just to change gears slightly now, Jenny, my specialism is in security and today at the keynotes one of your customers are on stage saying, hey, security is really important to us and even if you are a vendor that doesn’t specialize in security, that is a very big key pillar of what everyone’s worried about. So maybe keen to understand Oracle’s approach towards data integrity, privacy compliance. Given massive amount of data breaches happening out there, people are worried about it and data is really key. So keen to understand how Oracle’s approaching this.

Jenny Tsai-Smith [00:16:29]:
Right, so we start at the center of data. So the data that we’re protecting, we’re actually protecting them through policies, security policies that you can define on the data so that no matter how people are getting to the data, they have to pass through that gate, if you will, to make sure that whoever’s accessing it actually has the right privileges, the access control to be able to get to that data. So that’s really key, is just associating and putting that security policy right in the database where the data resides. So that’s step one. And then we protect the data also because we back up the data and we have a, a special kind of backup mechanism so that as soon as the data is committed, it’s backed up. So it’s not like a nightly backup or even an hourly backup, it’s as soon as the data is saved, it’s committed data, we back it up right away. And so in terms of the amount of data that you would lose, it’s pretty much zero because it’s backed up. The other thing too is the backed up data is such that you cannot tamper with it.

Jenny Tsai-Smith [00:17:41]:
So in terms of ransomware, right, somebody might want come in and try to tamper with your backup. Well, you can’t do it because it is also tightly protected in that way. So it’s a ransomware protection kind of feature that’s in our recovery appliance. And the recovery appliance is essentially the engineered system with some additional software to allow us to do that. And then beyond that we have the built in database SQL firewall. So if somebody’s trying to come in with a SQL statement that’s going to inject some malicious activity or try to take away data, we have the built in firewall again in the kernel of the database so that we can block those SQL injection attacks. And then going further, we have TLS 1.3 support. So we encrypt the data in flight.

Jenny Tsai-Smith [00:18:33]:
And in fact recently we announced the addition of using the NIST ML KM algorithm so that it’s quantum resistant. Because as you know, quantum computers, right, they people are working on them and they could be a very dangerous tool in the wrong hands. So we have that protection in flight. And then also we have of course the at REST protection encryption through our transparent data encryption and in fact all of our database cloud services, there’s no choice. The data is encrypted. Our customers who say, oh, we don’t want encryption, sorry, we have to protect your data. So that’s encrypted and that’s using AES 256 protection.

Karissa Breen [00:19:12]:
So when you say encrypted arrest. So there was a large data breach in Australia, big telco, and the issue was because it wasn’t encrypted at rest. But if it was, could that have completely been not a problem?

Jenny Tsai-Smith [00:19:23]:
Well, you know, I mean the data could have been taken, but because it’s encrypted and if you’re using AES256, it’s a quantum resistant algorithm, or as far as I know it is still, even if they have data, they can’t make use of it because it’s encrypted. They can’t really use it.

Karissa Breen [00:19:40]:
So just a quick side note on the quantum side of it. So I was interviewing someone recently and the term escapes me, but they’re saying once we have this level of decryption from quantum, they can actually go back and say, hey, if we stole that data, like we can actually decrypt it now.

Jenny Tsai-Smith [00:19:55]:
Right? And that’s what I’m saying is that we want to protect customers if they face that scenario. Right? So if somebody’s taken the data, we have to have encryptions that’s so strong that even if they have quantum computers, they wouldn’t be able to decrypt. Now, as I said, the encryption algorithm that we’re using for at rest data is AES256. And as far as I know, it is considered a quantum resistant algorithm.

Karissa Breen [00:20:22]:
Because that’s interesting because not only are people worried about what’s happening today, they’re now worried about 10, 15 years in the future that people can come back and go, hey, I’ve stole your data 20 years ago, now I can decrypt it.

Jenny Tsai-Smith [00:20:31]:
You’re right. Maybe 20 years from now the AES256 algorithm could be broken by some other more advanced quantum computer. I don’t know.

Karissa Breen [00:20:40]:
So then on that note, Jenny, what do you sort of see for the industry generally, or even for Oracle, what do you see sort of moving forward now, given what we discussed here today?

Jenny Tsai-Smith [00:20:48]:
There’s going to be more data, lots more data. And I think the goal is to allow people to get at the data without them having to move the data around. So we have customers who want something called 0 copy or 0 ETL or sometimes they call it Mirror Copy, but that sounds too much like replication. But the idea is that they want to be able to access the data and analyze the data institute where it resides, so you don’t have to pay for the extra cost of moving the data around, storing it. Again, you’re not duplicating data. So I think more and more that’s what customers want. And our announcement today with Oracle Autonomous AI Lakehouse, where we’ve brought broaden our support for Apache Iceberg, which is a very popular open table format that many other data management vendors support. That’s sort of the beginning of open data platform that customers can build on and benefit from.

Karissa Breen [00:21:47]:
And then lastly, Ginny, is there any sort of closing statement you’d like to leave our audience with today?

Jenny Tsai-Smith [00:21:52]:
So what I want to tell the world is Oracle AI Database has been around actually for longer than any other data vendor that I can think of out there. And so we’re time tested and yet we continue to innovate and we keep up and actually stay ahead of the trends. And so I would love for everybody to, if they’re not already using Oracle AI Database to come and try it out. And the fact is that we can support essentially any kind of workload that they want to run. So they don’t have to go and buy specialty databases to be able to meet their needs. They could come to Oracle AI Database and we actually have something called Always Free Autonomous AI Database. We have customers who are starting businesses like startups, they can actually use that to run their business and get started. So come and check us out.

Karissa Breen [00:22:48]:
Joining me now in person is Rand Waldron, Vice President, Oracle. And today we’re discussing Oracle’s position on public and private sector. So ran, thanks for joining me and welcome.

Rand Waldron [00:22:58]:
Thank you. Good to be here.

Karissa Breen [00:22:59]:
Okay, so I’ve had a look into your background. Quite interesting. And you’re obviously very well versed at understanding both of the sectors public and private. So I’m keen to understand the government’s adoption towards AI and typically it’s been more of a probably private sector historically that’s really adopted new technologies and are more willing to take risk. So I’m really keen to Understand your view, given your pedigree.

Rand Waldron [00:23:23]:
I think it’s a very, very complex set of constraints that governments have as they look at AI. Frankly, at any innovation, I’m not sure it would be all that different. Whether I were talking about AI or talking about the last series of innovations around cloud or any of those capabilities, what I am seeing is some real progress in government. And there’s a couple of different ways I’m seeing that progress. One, I’m seeing governments using AI to directly address their mission. So that might be citizen services, it might be a national security or defense mission, but they’re using AI directly against that mission set. Two, I’m seeing governments start to use AI to become more efficient in how they deliver those capabilities. So they may have a team that has a very long backlog of permitting or a very long backlog of addressing some veterans benefits or those capabilities, and they’re beginning to find ways to use AI to address that.

Rand Waldron [00:24:23]:
What I think is most fascinating, and this is where the kind of public and private comes together, is what we’re really seeing now is a generation of very exciting, young, innovative companies that are being born to address these public sector problems. And so what you’re seeing is you’re seeing people who often they were in the government, maybe they were a veteran, maybe they were in the intelligence community or in the civilian side of government. And there is a particular problem set that is really a passion project for that individual. There is some particular problem that they were not able to really solve when they were in government due to any number of issues. But what they’ve realized it is that in the world of AI, they can actually build a technological solution that solves that problem. And so they’ve begun to create these really innovative companies that often have or start with very near narrow niche of a particular problem that they want to solve that they experienced in government. They’re using AI, they’re using cloud technologies to solve that problem, and then they’re working back inward with their former government compatriots and bringing that solution into government. And so I don’t want to over abuse terms like public private sector partnership, because that’s not even really what this is.

Rand Waldron [00:25:48]:
This is small companies solving niche problems that they care a lot about, using AI to do it and then finding ways to bring that back into the government agencies. So I’m seeing that really globally, in Australia, United States, in Europe. I think it is incredibly powerful and it is something that would have been very hard to do before the world of AI, because solving a Niche problem like this would have taken too many people, too much investment for a niche problem. AI allows you to do it at a much larger scale with a very small initial investment.

Karissa Breen [00:26:20]:
Because historically speaking, people probably wouldn’t associate like government agencies doing like innovative, sort of leveraging technology and AI like we’re doing today. So it’s more like is is the perception then changing that government is moving more like a private sector would or.

Rand Waldron [00:26:34]:
So I don’t want to fall too deeply into that perception because I actually think that government agencies can be and have been quite innovative, right? So we can go back 50, 60, 70, 100 years and a lot of the innovation that we stand upon today originated in government research and originated in government agencies and all of that. So I don’t want to buy, buy fully into the stereotype that governments aren’t as innovative. But I will say this. Governments have a series of equities and stakeholders that are really important. I always tell my people that it’s the world’s most important work, is the work of the governments. So they do have these stakeholders, they do have these equities that really are deeper than kind of your average company. And so it is difficult for a government agency with these kind of critical stakeholders, with the mission critical work they’re doing to experiment, to just try things. If you’re a government agency, you can’t move fast and break things.

Rand Waldron [00:27:35]:
You’ve got to move fast, but you can’t break anything. And so what I’m seeing is more and more government agencies looking for ways to use these small companies as ways to move fast, as ways to bring innovation into their agencies. And they don’t want to break anything, they can’t break anything. But that innovation can happen without the break things side of the equation. So I am seeing that innovation coming. I’m seeing it still what I’ll call very personality driven, meaning you’ll get in government somebody who is very driven by the idea of innovation. And she will focus her organization on innovation. She will focus her organization on working with some of these private sector companies.

Rand Waldron [00:28:19]:
She will pick a few places where AI can scale her organization. The challenge is that often an individual like that ends up moving on after a couple of years. And so she gets promoted, she moves on, she goes back out to industry, whatever it ends up happening. And it is difficult to get that innovation stream institutionalized into the agency rather than based around her individual personality. And so that’s what I’m seeing. I’m seeing individual personalities very driven by the gains that their agency and their citizens can get from AI. They’re working with small companies to do that. But it is still a challenge to institutionalize that rather than have that just be personality focused.

Rand Waldron [00:29:03]:
Does that make sense?

Karissa Breen [00:29:04]:
Of course. And would you say that that’s going to change over time? Perhaps, or.

Rand Waldron [00:29:07]:
It will definitely change over time. There is no alternative. Governments have essentially an infinite number of requirements and a finite number of resources. And the thing that bridges that gap is software, like humans do not scale well. You, there’s only so much I can do. I only have a certain number of days, hours in the day, days in the week, weeks in the year. But AI agents and AI capabilities scale infinitely. So you may only be able to hire a certain number of humans to solve a problem like approving permits or better managing veterans benefits or something like that.

Rand Waldron [00:29:43]:
You probably only have authority to hire a fixed number of humans. If the problems those humans have to solve is just a longer list than they can handle, then we end up with backlogs. It takes forever to get permits approved. Veterans benefits take a long time to adjudicate all of these things. But AI scales infinitely. So you can then say, look, this is the particular problem we’re encountering and can we build AI capabilities around solving that? The answer is absolutely yes. And governments are being forced. And forced is not even maybe the best term for it.

Rand Waldron [00:30:18]:
Governments are now seeing the possibility of this. So you can take your current team that is doing everything they can and still has a backlog. You can work with an innovative company, whether it be Oracle or a small company, you can work with that innovative company, they can build technology with you that helps you get through your backlog faster, and citizens end up with better service, better capabilities. And if you don’t do that, your only alternative is to go to your government, your legislators, and ask for a bunch more money. And I can tell you around the world the answer to that is no. And so you, you governments will innovate through this problem because they have to innovate through this problem.

Karissa Breen [00:30:59]:
So can you share maybe any sort of recent examples of US Government agencies using OCI or for AI successfully that you can sort of talk through?

Rand Waldron [00:31:07]:
Yeah, absolutely. One of the things we’re proudest of is the work we’re doing in Oracle Healthcare with Veterans affairs and the real capabilities we’re bringing to make life easier and more efficient for both the care providers and also for the veterans, the recipients of the care. And so there are aspects that we are building around better patient portals, better access for patients to understand their care journey there’s capabilities we’re building into systems to allow care providers to use things that might sound simple like auto dictation and auto filling out of forms and all of that. But what you see is that that seems like a simple thing, but it drives a lot more efficiency. And then done right, it helps actually avoid missed problems and all of that. So we are seeing it in care scenarios like that. We’re also. I was just talking with a small innovative company out of the United Kingdom who’s working with the Royal Navy on bringing AI into their world of understanding.

Rand Waldron [00:32:11]:
Lessons learned. It’s one of these things where parts of our military forces around the world are constantly. They’re going on exercises, they’re working with new equipment, they’re trying things out. And in doing that, they learn a lot of lessons, they learn a lot of new capabilities, they learn what works and what doesn’t. And all of that gets recorded and written down and tracked and managed and. But is anybody generating insights out of the thousands of pages of lessons learned of things that we understand about this particular capability, how it worked in this particular exercise? All too often these lessons learned are being generated and recorded, but they aren’t turning into insights that then drive the next set of innovation, because there’s simply more information and more data than these organizations can absorb and communicate into their next wave of innovation. This is a perfect AI problem. You can have AI look at those that data, summarize it, understand it and communicate it, rather than as a hundreds and hundreds of pages of documents communicated as actually action items that then other people can innovate against.

Rand Waldron [00:33:23]:
So we’re seeing this throughout the.

Karissa Breen [00:33:25]:
So basically you guys have got the horsepower to be able to analyze all of that and then derive insights as like, hey, this is what we know out of pages and pages and pages.

Rand Waldron [00:33:33]:
And what I will tell you is that no one company, even one as big and as innovative as Oracle, no one company can do this all. It just isn’t possible. We’re building incredible health care technology, we’re building technology for public sector and law enforcement. We have business units working on thousands, tens of thousands of problems. But there are millions of problems. And so the real answer is to say, look, how do we, as Oracle, innovate as fast as we can, but far more importantly or just as importantly, how do we build a platform for all of these other companies to innovate against? How do we provide a set of capabilities, whether it be through generative AI or whether it be through some of our kind of more classic AI capabilities? Around facial recognition, around translation and all of these capabilities. How do we provide through our cloud, the infrastructure that in the past would have only been available to a billion dollar company, and how do we make that available to a small company innovating against a particular problem that is passionate for them, that they are deep subject matter experts in, and we provide a platform that lets them innovate against that. That is how these problems get solved at scale.

Karissa Breen [00:34:56]:
So I want to slightly shift gears now. We’ve seen many breaches, outages, et cetera, in the US but broadly across the world, as you’re obviously aware. So a couple of words for me come to mind. So transparency and accountability. So how does Oracle ensure transparency and accountability towards AI and cloud are leveraged within the public sector? Because it’s something that we, when I’m doing a lot of reconnaissance or research, even if I’m on Twitter, I mean slash X. Just the sheer volume now of people holding these companies accountable, we’ve. The loyal loyalty probably just isn’t there like it used to be, right? You know, we don’t go down to the same bank and see the same person that we did the banking with. People are quite happy to move.

Karissa Breen [00:35:34]:
So I’m really keen to understand how that looks in your eyes.

Rand Waldron [00:35:38]:
So I think that AI brings the chance to bring accountability and transparency at real scale. And this is something that, believe it or not, I am seeing governments embracing rather than pushing back against. And so the AI brings now the capability for governments to share masses of documentation with the public and for the public to actually understand all of that. And if you ask your average citizen, well, there was a Freedom of Information request, it generated 10,000 pages. I know you have a job and a family, go read those 10,000 pages, right? This is not a practical way of delivering transparency. Flip it the other way. You have citizens who deliver a lot of feedback to their governments. And the governments sometimes are overwhelmed by this feedback.

Rand Waldron [00:36:32]:
Digesting that feedback can sometimes take months or years and make projects take even longer than they need to. What AI does is AI can allow both the government agency and the citizens to better communicate with each other because they can digest each other’s scale. It’s very hard for a citizen to truly understand the scale of what governments do, the scale of the information that governments create, even the scale of information that governments release. AI gives citizens an ability to understand that. I’ll talk about it in the us, right? Like we create a budget document. Every year that budget document gets approved, or as right now, not approved, and how many Citizens actually understand what’s in it. How many citizens actually read it? Is it, Is it zero? It’s pretty close to it. And you think about that’s one document.

Rand Waldron [00:37:24]:
How much is the government generating at any given moment? Essentially an infinite amount. AI will provide a window for individuals to understand their governments in ways they cannot today. And I think that there is tremendous potential for transparency there. Just as an individual, I look forward to that transparency. I look forward to the ability to understand what my government is doing by better having AI help me see it. And I can create some of that traceability. It can tell you this is something I found in these documents. And by the way, this is where it is, so you can actually go read it in the actual, actual document.

Rand Waldron [00:38:03]:
And so that’s, that’s really where I see AI helping us create transparency.

Karissa Breen [00:38:08]:
And do you think as a byproduct of that, it’s going to sort of close the gap between just, hey, I’m a Citizen, one of 341 million people that live here, but my needs are being addressed? Because you can actually read this document, that makes it make sense. Because a lot of the words and the terminology is probably the vernacular may not make sense to everyday sort of person.

Rand Waldron [00:38:27]:
Absolutely. I know that something I have to work against after a career in government, there’s almost an entirely separate language that is second nature to me, that if I even try to communicate it to my family, if I tried to talk to my family the way I did when I was working inside the government, it would literally be like a foreign language. They would not understand it. They don’t need to. Their lives are involved in different things. AI can absolutely help bridge that gap. But I also see it in another way when, when you look at citizen services, one of the most common concerns that citizens have is right along what you were saying, which is they, as an individual don’t feel sufficiently addressed, sufficiently heard, or sufficiently served by their government. And I will tell you that is not because the people in the government are evil or lazy or what have you.

Rand Waldron [00:39:22]:
It is because there is more need to serve citizens than there is capability from the government to do that serving. That’s just the case. That is one of the big reasons that citizens feel like they have to wait long times for things like permits or driver’s license or whatever particular thing you might talk about. AI will allow governments to accelerate service delivery and make that service delivery feel more personal to the individual citizen. And that I think, done right is going to help bridge this gap that is between the Governments and the people they serve, I actually think there’s tremendous potential there.

Karissa Breen [00:40:02]:
So it’s a fair assumption to say, when they say, can we do the survey at the end of this call, people are going to start probably doing it.

Rand Waldron [00:40:07]:
I am not sure of that. I don’t think I’ve ever done one of those surveys. But what I’ll say is this. If in the past you had to wait 30 days for the government to accomplish something for you, I. And I don’t know what that is, that could be a permit, that could be a new passport, what have you. I deeply believe, because I am seeing it now, that that will be able to be delivered to you not in 30 days, but in a day or two, maybe less. And that you’ll be able to get it in a way that is more personalized to you and in a way that feels more like a direct relationship with your government and less just like your One more task on the list of a civil servant’s infinite list. And so AI will make governments feel more responsive and more personal to citizens.

Rand Waldron [00:40:58]:
There’s tremendous potential there.

Karissa Breen [00:41:00]:
So just to sort of come to the end of our interview, I’m keen to maybe just understand budgets. People want to do more with less. They’re expecting more from vendors. They want performance. They want all these things, all these features and functions. Help me make sense of this now.

Rand Waldron [00:41:15]:
From a public sector perspective. And this is something that I love about the public sector, whether it be on the civilian side or the national security and defense side. And it truly is that the problem set that the value that a public sector agency can deliver is essentially infinite. We talk about in the commercial sector, like, how big is your total addressable market? How many people can you actually serve with whatever product you. You invented and you’re selling? And you can put a number on it, you can say, well, I sell this product. It’s a somewhat niche product. There’s only this much demand for it. That is not the case with government agencies.

Rand Waldron [00:41:52]:
Government agencies have essentially an infinite amount of demand in front of them, and they have a limited set of resources. The only way that they are going to square this circle is by deploying technology. There’s all sorts of different parts of technology, right? There’s the cloud, which we’ve been talking about for years. There’s capabilities at the edge to actually understand things throughout a city or in the field for a military and defense organization. And there’s the AI capabilities. We’ve been talking about so much there. The reality is, with an infinite mission in front of you. A government agency has really three choices.

Rand Waldron [00:42:28]:
They can agree to over time deliver worse and worse service. They can go back to their funders and request more and more money. Or the third option is they can find a way out of that trap and they can use technology. They can use AI to deliver a lot more service for their same or fewer resources. What I want to say is like that is somewhat a bland thing to say because yeah, everybody wants to do more with less. But what we’re seeing today, what is so unique about AI is that AI is actually doing it. You can actually create AI agents that reduce the time for a permit review for an individual. Permit reviewer has to review a tremendous amount of documents and it might take them a week to literally just read and understand those documents.

Rand Waldron [00:43:20]:
And the AI can reduce that significantly, can allow the human to interact with the documents rather than as an individual reading their way through a stack of documents. The AI can allow that reviewer to go to what’s relevant, what’s critical and what’s important and radically reduce the amount of time that they’re using. Reviewing that document. That does not reduce quality. They’re achieving the same quality or better quality, just with a lot less time because the AI helped them do that review faster. And that translates directly to that citizen getting his or her need addressed faster. That is a practical thing that is happening today that innovative companies like Oracle, like partners we work with, are developing capabilities against. Today that I think is the only answer to the budget problem.

Rand Waldron [00:44:10]:
There isn’t going to be a world where across the entire democratic world we’re going to be able to go and ask for infinitely more money every year. It’s just not going to happen. We also cannot have our governments become more and more distant and feel less and less responsive to their citizens. AI bridges the gap. They’ll be able to do more and feel more responsive to their citizens.

Karissa Breen [00:44:32]:
And lastly, Ran, do you have any sort of closing comments that you’d like leave our audience with today?

Rand Waldron [00:44:37]:
I’ll leave it on that hopeful note that there is no one company that will solve this problem. Oracle is doing a tremendous amount of work to help, but there is this really hopeful world of companies that are being built small. They’re not all small anymore, but small and growing to address really specific problems that their founders are passionate about. And we’re creating a platform that lets them do it. And there are things that we are deeply passionate about. Right. We spent this week a lot of times speaking about healthcare and applying AI to healthcare. And we spent a Lot of time this week talking about how we’re creating these absolutely hyperscale AI capabilities.

Rand Waldron [00:45:23]:
The intent of those is to make more technology and more capability available to each of us as an individual and to the smallest companies. And that capability will allow us to solve the biggest problems and most important problems in the world. And those are the problems that my public sector customers have.

Karissa Breen [00:45:49]:
Joining me now in person is Arman Ashouriha, head of Oracle Cloud Infrastructure Modernization Program at Vodafone. And today we’re discussing Vodafone’s journey to modernization on Oracle. So Aman, thanks for joining me and welcome.

Arman Ashouriha [00:46:00]:
Thanks for having me.

Karissa Breen [00:46:01]:
Okay, so Aman, I’m aware that you’ve been a long term customer of Oracle and recently I’m keen to really understand what was the catalyst for taking this relationship into a full blown sort of modernization program.

Arman Ashouriha [00:46:13]:
Pretty simple actually. So we are very heavy loaded Oracle users and long term customers. So we are hosting majority of our services on Oracle databases across the the European markets and even the African markets. So and we had a big challenge and the challenge was end of service life for a large state of our Oracle database. We are talking about large, large numbers. And for us to go down the route of even strengthening and deepening our partnership was to find a solution which will enable us to do mass migration on a factory base across a three years program to protect our business. Because a lot of these databases there are even Oracle applications sitting on top which where our customer data set sits on. We have to protect the business, be compliant and then avoid basically technical debt.

Arman Ashouriha [00:47:04]:
And for that reason what we did, we evaluated internally technical options and then together with Oracle to find a solution how we at scale can do modernization of our databases and some core Oracle applications. And with that Oracle is the only company which offers you basically a public cloud, but on premise, the regional cloud cloud, they have the OCI regional dedicated regions which we use and we are one of the largest customers in the world. We deployed six times the drcc, the dedicated regional cloud health customer solution, two times in Germany, two times in Italy, two times in Dublin, Ireland which are our main data centers. And the reason why we went for that solution, it’s a full blown public cloud. I have it in my data center, it allows me actually to do it if needed, a two steps modernization approach, meaning separating the application, having the application, meaning having the application in a data center in the traditional world and then having the database also in the same data center but in another room or in another house on the new modernized Oracle cloud infrastructure so this enables you first no latency challenge because we are very dependent on latency. If I would move now, one part try to move into public cloud and the other part would be on premise that will not work in the majority of the telco services. So that for that reason we brought it in house, separated it and migrated and modernized towards the Oracle. So that was actually the key core because we wanted to have an acceleration approach to be able to do multi thousands of database modernization.

Karissa Breen [00:48:39]:
So just sticking with the latency challenge, would you say given your role, there is a lot more pressure now on businesses to perform constantly for their customers because customers expect a lot more nowadays?

Arman Ashouriha [00:48:52]:
Absolutely. I mean there is a fantastic. Actually it’s a very, very good question. For example, if I take our UK market, which is our second largest market in the uk, which is very different, for example our biggest market is Germany, but in the UK the iPhone launch day is the day where you’re going to decide on the market who wins the year with the competitor. And for us performance and especially using, we are using a lot of our customer related applications are all on Oracle in the UK and that day it’s all about performance and operational stability. Performance from a user experience, someone who access and our call agents, our stores where a customer wants to sign a new contract, renew it. It’s all about experience and performance. Yeah.

Karissa Breen [00:49:37]:
Okay, so that’s interesting going back to what you’ve just said before as well, would you say, and you mentioned before we had to make this decision with Oracle, but internally, would you say as well, given the current climate of companies having that competitive edge of their competitors, we have to start to make decisions more like faster nowadays as opposed to even five, ten years ago when we’d have a very long risk assessment. We get all these people in the room. Would you say even at a business level the decisions are being made faster than historically speaking?

Arman Ashouriha [00:50:07]:
Absolutely, yes. Because even this initiative which we run, so we had to make a fast decision across 19 entities, which is a very difficult thing and especially across 11 countries. So and we assessed, I was in the process of the assessment from a technical then to the commercial part, nine months. I think the overall was, I think around 15 months where we evaluated, once the decision was made, within 6 months we builded the first region full blown cloud in one of our data centers. Six months later we bought one through one night, five cloud regions live and already migrated stuff on it. So time to market is the key actually. And operational stability as much as you can, zero downtime. If it’s possible for the critical applications.

Karissa Breen [00:50:55]:
And you said before zero downtime. So one of the things that I’m hearing a lot in the interviews that I’m conducting amlan, would be continuous business. I don’t know whether you’re an Australian or my accent is Australian. And recently in Australia, one of the telcos had multiple issues, but again it meant that people couldn’t even call out. So it was more so their downtime has caused significant problems. And would you say even looking at other telcos, that is giving some awareness of what could potentially happen with that downtime?

Arman Ashouriha [00:51:24]:
Absolutely. Now we need to distinguish between one is a network downtime, because in the telco, as a consumer, one thing is, do I have connectivity? Right. Another thing is for a customer like yourself, you might use an app or you log in from your email, your laptop and you want to do changes on your contract and you cannot. That’s where it comes in. And I’m representing the IT part. The most important thing is you’re in a very competitive market, right? You need to distinguish service quality, added value services and value for money and user experience at the end.

Karissa Breen [00:52:00]:
And so then these are sort of my next point around technical debt. So I’ve come from a banking and finance background quite across, you know, legacy systems, et cetera. So maybe I’m really keen to understand what that sort of meant for you in the Vodafone scale and setting.

Arman Ashouriha [00:52:16]:
So very simple technical depth. Unfortunately, we have that challenge extremely as well. And it comes to various reasons. One is we grow through acquisitions and sometimes merger and acquisitions is one thing. And then we have also joint ventures in some markets. You basically, you need to run multiple, even customer based systems for a long period of time before you can even migrate them, merge them, centralize them to the Vodafone standards. And for that reason, so not only that you’re facing your own old applications, you’re trying to go to the modernization lifecycle and et cetera, you inherit even more. And then the challenge comes on top of the technical depth is standardization.

Arman Ashouriha [00:52:57]:
Right. Because in an ideal world I should have one platform centralized from a group point of view and then have local content on it, rather than having for every single market its own platform and its own content.

Karissa Breen [00:53:11]:
So then you mentioned before you got all these regions that you look after, so how do you sort of then roll that out?

Arman Ashouriha [00:53:15]:
So the rollouts, that was one of the most important aspects. So what we did with Oracle was there is a process as a huge company which we are used to, we put it all aside and Said, okay, look, we are on a mission. When you’re on a mission, all what matters is actually always having the business goals ahead and then see if there are obstacles, how are we going to manage the obstacle and how much of a risk can I take? So what we did, we had to prepare physically data center space, because in this solution we provide the data center space, the power, the cooling and Oracle comes in and builds its full blown cloud solution where they own the asset, which is a very positive thing because I don’t need to take care anymore about end of service life and life cycle management of the asset. I only need to take care about the application then in the future, which is a fantastic thing for me because I got rid of that problem.

Karissa Breen [00:54:05]:
Okay, I want to get into this a little bit more because I think this is something that dealing with large enterprises, everyone’s got technical debt, legacy systems. What would you say going back to your comment around the risk side of it, how what were some of the. Can you high any sort of those risks or what was going through your mind?

Arman Ashouriha [00:54:21]:
The biggest risk you need to take is because you’re going to touch running systems, right? So one thing is you need to convince the respective markets which have a lot of pressure, business pressure coming from the CEO in the respective challenging local market. So one thing you need to do is basically convince people for change, which is a big, big subject because a lot of these big transformations where modernization is part of the first thing is not technology, it’s the people. These are the persons, everyone involved who needs to have a skill change, a mindset change, then you can run through that. So the challenge itself came with that. We had to create a complete new way of working with a new spirit, with a new team approach and be partnered into the lowest level you can imagine with Oracle and the same time at the highest level having direct access to the XCO of Oracle from our side. So we worked as one team and represented us as one team. There was not. You are a vendor.

Arman Ashouriha [00:55:20]:
I’m expecting this because we had even we had always in a rule in our executive staircase which were every week we were not allowed ever to mention the word contract because move that all the way. Because if we are on a mission all what you care is how do I convince the markets to be part of this? They know they need to do, but are they hesitant? And another thing is giving them trust and operational performance engagement. And then later once we migrated very complex application and databases and they saw the huge benefit of performance improvements and operational stability.

Karissa Breen [00:55:54]:
I’m going to sort of talk to you a little bit more about why this isn’t just a migration, but more sort of a reinvention. Like I’m really keen to sort of unpack what that means in practice.

Arman Ashouriha [00:56:05]:
Very good question as well. One of the biggest product from a product perspective change is once you move to a public OCI or this Oracle dedicated cloud, which is the drcc, which is as I mentioned, the same product services which you have on a public oci, you finally get access to adb, the autonomous database. Or the autonomous database. We are using large scale exadata in our traditional words. But to avoid in the future that I run into similar challenges of lifecycle management and then again patching and et cetera, the autonomous database, which is a proper transformation, it helps me up to roughly 80% less human interaction and maintenance on a running business. So it does automatic patching. One of the challenges I came with, one of the previous sessions I just attended was also how do I convince my internal customers that they go with me on this transformation journey? Because it’s a different product, it brings all the capabilities, but it’s way beyond. And people, as you can imagine, always want to control things, but now you come and it does it for you.

Karissa Breen [00:57:12]:
So because people want to control things, how is that sort of how people thought about perhaps relinquishing some of that control? You said before, 80% of it’s being.

Arman Ashouriha [00:57:20]:
Automated now as I mentioned change people are always redundant right towards it. So what we basically did, the best way to do it, you start some with some application services, you own Internet where you can take decisions by measuring the risk. And I said okay, this works, let’s test, let’s do with some of our markets which were very willing because they had pressure. Because then the autonomous, if you use it properly, you reduce your tco, which is very, very important as well. And then you gain trust because all what matters is suddenly you run an autonomous database and you literally see the benefit of it and then the cost goes down and operational performance is given. And by doing experimenting to be fair, because if you don’t know, you need to convince people and show them the showcases and the improvement of performance. Because even for an OPS person who’s used to access and work in a specific way, one thing you are always convinced with the time you require to log in, if that decreases heavily as well, people are already one step towards the change, which helps.

Karissa Breen [00:58:24]:
So when you sort of look back, is there anything that people should know that are more nuanced knowledge that you have now after going through something like this, would you say?

Arman Ashouriha [00:58:31]:
Absolutely. I think doing a large scale program across so many entities. As I highlighted before 19 what matters is you always should have your business objectives as a goal rather than the technology to be fair because it’s the people who makes it a success or not. Technology is just an enabler for us. So for me is the preparation to understand what you’re targeting then how much of a risk you can take. The planning is important. The governance aspect. Working with so many complex entity, the network, the connectivity is a big challenge for old systems because for majority of them you need to trace back the connectivity matrices because you don’t have them anymore.

Arman Ashouriha [00:59:13]:
So we found ways to kind of standardize at large scale to do that. Do discovery back on your old database trace. Okay, what is it connecting to? Then at the end, preparation, change management plays a role, big role. Having security as part of it from day one because you have a public cloud on premise which in theory you can use as a like a public cloud because give access to Internet or you use it as an internal cloud. So all these aspects. All that matters is a proper preparation in terms of what is your mission, what are you trying to achieve it and then having the right mindset and dedication of people. Dedication of people means really give them the tools and mandates, the trainings, the learning on the job. Because it was a brand new technology.

Arman Ashouriha [01:00:00]:
You had to learn and grow at the same time. And management support, literally management support.

Karissa Breen [01:00:06]:
Just before we move on. I’m keen to maybe I worked on the tail end of core banking modernization for a bank. And that was you mentioned before, risk, good risk management, good planning. What about the execution side of things? I think programs don’t run over budget or they go for a protracted amount of time.

Arman Ashouriha [01:00:23]:
That’s a very well questioned and that’s typical also in our case normally. So what we did here, it was a program which was never planned, no one planned. That’s why we had to form a group. Because I sit in the group entity, not in a market. So we went with a plan and went through all the single market CIOs, CFO, CEOs to make them understand what the risky is they are all the time taking. And then we said okay, we do a dedication, we do from a budget point of view, a long term planning for a duration of time. And this budget is set, fixed in stone. And then we went on the other side is together with Oracle we created a massive factory.

Arman Ashouriha [01:01:01]:
And I have to admit I got the target of modernizing through this period of 60% of our European estate. And we had the program split in two phases. Three years with Oracle and the last year which we just started 1st of October, actually focusing internally taking over some of the methods we created the standardization, we created blueprints, pipelines and everything. And we internally as Vodafone, we carry on and I have to say 12 months now ahead we are we touchpoint and modernized 64% already one year ahead. That’s why we’ve been pretty successful in what we have been doing actually. So it has been a very successful program. And now what we are going to do as a side evolution of it, you’re going to use the multi cloud solution, using even Oracle at Azure as an example, as an X phase and using a lot of the learnings and methods standardizations which created even for that.

Karissa Breen [01:01:55]:
So then I’m curious to know how has OCI sort of changed the speed at which you can deliver sort of like new services now would you say?

Arman Ashouriha [01:02:03]:
So far immense. I would say because it starts Oracle launch a new service On a public OCI maximum latest 6 months later I have it automatically in my catalog which I can use and deploy. So if I would need to try to create and develop these services myself, for example automation services, orchestration services, then I would need to assign people dedicated roadmap, give them dedicated budget. Now what we do is we have a rule out of the box first. So whatever Oracle provides for monitoring, security, support and et cetera, let me use the native solutions which Oracle is developing for me and making available for me. Then I have to say does it match to my compliance and my standards as a company? And if not, in a lot of cases it does already. I just build on top of what I need to be compliant to my Vodafone, in this case Vodafone stuff for the other regions.

Karissa Breen [01:02:55]:
You mean then as well, because you should have been. Okay, got it. So I want to slightly change gears and talk about migrating off a platform like MongoDB, which isn’t trivial. So what was sort of behind the decision? We’ve sort of spoken about it a little bit, but keen to understand more about the advantages in making the move.

Arman Ashouriha [01:03:15]:
Perhaps we are still at the early stages. We did just a few and the idea came from. So the funny thing is when we set up our program, we call it ocimore and the more has a double meaning modernized Oracle estate and more, literally more. And for me was when I look into our estate super large, we have also Mongo we Have some postgres, we have some other type of databases. What I was keen is how can I optimize and standardize my strategic products. So and my strategic product when it comes to database. Absolutely. Oracle.

Arman Ashouriha [01:03:48]:
And then if I we started from a logic, what if I try for applications I own as an organization that I’m based on, see if it works, if I might move this from Mongo to Oracle. And the reason is super simple. I don’t need a dedicated team which is also taking care of Mongo. I can save costs and at a larger scale if I look across the whole Vodafone it would give me a lot of benefits because standardization and the new capability because we work also with Microsoft, which Oracle gives me as well is not just replacing Mongo. And through the AI capability they are embedding in this 23 AI now they launched announced yesterday the 2026 AI. I have way more service available which is through the AI mechanism they have than I have on a Mongo even one thing is saving another thing standardization. The third part is creating added value for application development on top.

Karissa Breen [01:04:44]:
And so what would you sort of think now moving forward? How do you sort of see, I know you’re sort of saying that you’re in like the next sort of phase of it, but what do you sort of think moving forward, how is this going to transform what you guys are doing at Vodafone?

Arman Ashouriha [01:04:56]:
For us, so the next step is through the drcc. We enabled hybrid cloud as I meant touchpoint. So we are using like a private cloud, but in some specific UCs, like a public cloud. And we are connected to the public cloud because a lot of our applications are speaking to each other which are already in AWS or wherever they are. And we have the absolute strategy public cloud actually first, but then we have critical systems maybe regularity or the nature of the application doesn’t allow to move it to a public cloud. That’s why this DRCC solution is so important. And we are using now Oracle as a multi cloud. So as I mentioned, vast majority of our applications are using Oracle technology if they are not already an Oracle application itself.

Arman Ashouriha [01:05:41]:
So our strategy is public cloud but are using Oracle across meaning just we are waiting now in specifically in Europe that Oracle is going to go live I think in two months or something with the Oracle at AWS even gives us another option because we have large workloads already in aws. Then I can also have my standardized Oracle performance database also as a supported service also even there. So Multicloud is the Absolute direction.

Karissa Breen [01:06:05]:
And then just lastly you touched on a lot about performance. Would you say given your role like that? That’s a very big focus now because as you said to maintain competitive edge with all the stuff that you guys are doing in terms of the workload, et cetera. Like performance is what a lot of people are critiquing these hyperscalers on.

Arman Ashouriha [01:06:23]:
So I think on the performance side there are two types of performance that matter for vast majority of when we specifically when we migrated the database to and the application we got a lot of feedback that up to 70% performance increase from a let me say database admin or an ADB ad which is the application data based operations people who are accessing the systems in terms of performance. The third one if it’s a customer facing one, the performance we measured in some applications we improved by 35 to 50% the time which transactions are happening at the cost end user side and that’s what incredible. Even from an infrastructure point of view if you take an old database you don’t even modernize it putting it to cloud you see huge performance input because you need less CPUs because you’re running on a way, way more modern X11s systems and ET cetera. So there are multiple ways where you see massive performance improvements and aman.

Karissa Breen [01:07:23]:
Is there anything that you’d like to sort of to close any sort of closing comments final thoughts you’d like to leave our audience with today.

Arman Ashouriha [01:07:28]:
I think we are all on the same journey especially if you are large traditional companies which are existing since many many years. I think the most important thing is that’s why such events are very, very helpful is we create communities because we were one of the early adopters on this DRCC and then we created together also with Oracle we’ve been customer references and you learn from each other when you speak to other people not even in the telco industry. They are in the same challenges which we have. They have them now and the next phase is now making properly use of AI of course. Right. So this community I think is. Is a massive, massive focus which I personally personally am putting on is learn. Let’s learn together and share together the goods and bads.

Karissa Breen [01:08:13]:
And and there you have it. This is KB on the go. Stay tuned for more.