How Organizations Can Get the Best of Oracle Database and Google Cloud

Oracle Database Infrastructure + Google Cloud = efficient deployments? Host Patrick Moorhead is joined by Google Cloud‘s VP & GM of Databases Andi Gutmans and Karan Batta, SVP at Oracle Cloud Infrastructure for a conversation on the synergies between Oracle Database and Google Cloud, and how organizations can leverage these technologies to enhance their data management and cloud strategies.

‍

Highlights of their talk include ⤵️

‍

• The unique advantages of integrating Oracle Database with Google Cloud services
• Strategies for seamless migration and deployment of Oracle workloads on Google Cloud
• Considerations for maximizing performance and security when using Oracle Database in a Google Cloud environment
• Success stories and use cases of businesses utilizing this partnership
• Future developments in the collaboration between Oracle and Google Cloud

‍

Learn more about Oracle Database@Google Cloud here. Watch the full video above, and be sure to subscribe to our YouTube channel, so you never miss an episode.

Patrick Moorhead:
The Six Five is back and we are talking about one of my favorite topics, and that’s multi-cloud. About 10 years ago, my research firm and I, we talked a lot about multi-cloud. We were called cloud deniers. That’s a funny thing because I used to provide infrastructure for the largest cloud providers out there. But I understand it takes many forms, but the enterprises have spoken and they want to run their cloud wherever they want, whether it’s on-prem, whether it’s in the public cloud and even having industry leading databases in their favorite cloud provider. And with that said, I’d like to introduce Karan and Andi to the show here. Guys, great to see you.

Andi Gutmans:
Thanks for having us.

Karan Batta:
Thanks for having us.

Patrick Moorhead:
Yeah. So listen, I can’t imagine the two companies to me that exemplify multi-cloud are literally Google Cloud and Oracle. While a lot of folks were hugging it, “Hey, we don’t want data in. We don’t want data out. We only want our flavor of our stuff to give you access to.” So I give the two of you a lot of credit at Oracle and Google Cloud for doing it. But hey, there have been a lot of announcements, I go to Google Cloud Next, I go to OCW. Let’s get us caught up on what’s going on between the two of you. And maybe we’ll start with you, Andi. Can you talk a little bit about some of the new capabilities? What regions did you announce with Oracle database at Google the last couple of months?

Andi Gutmans:
Yeah, absolutely. So as you said, openness has always been the core part of what we believe in at Google and really enabling our customers to adopt a multi-cloud strategy. So the partnership with Oracle just made a ton of sense. Obviously Oracle database is everywhere in the enterprise. And so our customers really wanted us to work together and be able to use the best of Oracle and the best of Google together. So last year we announced a strategic partnership to basically enable our customers to leverage Oracle databases within Google data centers, and also be able to run Oracle applications on top of Google Cloud. And that’s just been received in an awesome way. We announced four regions last year as part of the announcement. And just recently we announced the forward looking availability of eight more regions that are coming, which we’re really excited about. And I would say that’s the minimum amount of regions we’re going to launch this year. We’re actually hoping we’re going to launch even more than that, so at least a total of 12 regions. And that’s really a great sign for the demand that we’re seeing from our enterprise customers for this joint partnership. And that includes services like Exadata Database Service, Autonomous Database Service, and there’s more cooking as we’re going to bring more of these technologies together.

Patrick Moorhead:
Yeah. So Andi, just so I can educate or elucidate folks either watching or listening, essentially those customers with an Oracle database could really activate the data, let’s say on a Vertex AI or something like that. Am I reading this correctly? I think I know the answer to that, but I’d like to have you explain it.

Andi Gutmans:
Yeah, absolutely. One of our big strengths is AI and how Google has innovated in AI helping customers activate AI. Now AI, these foundation models are great, but if you don’t bring the enterprise data to the foundation models you’re not really able to build true gen AI applications. So this is really enabling customers to bring their enterprise data to where AI sits. And of course, we have additional services like analytic services and other exciting services they can take advantage of. So it is definitely an ability to converge their database estate and data state and bring that closer to where they’re doing a lot of their AI innovation.

Patrick Moorhead:
Excellent. So Karan, I want to bring you into this as a former product guy which we talked about in the virtual green room. I have a team of analysts, but I like to dig deep into the bigger ones. And this was definitely one of them, everything you’re doing with Oracle database. And I look at, hey, is this a paper launch? Is this the press release? Okay, we’re going to be GAA in like a year, maybe it’s going to be one region. I know Andi already talked a little bit about how much is out there, but you announced eight new regions, 12 in total in the next year. Maybe talk a little bit about the status of those, but also what’s driving all this increased capacity. I tune into your CapEx, what are the numbers in your earnings call? I always hone in on, Larry gives the CapEx number, but can you talk a little bit about that?

Karan Batta:
Yeah, absolutely. I think if you think about the velocity of what the two companies have done, it takes really two to tango in this case because it is a partnership. So we announced this thing I think within a week we sort of announced it and then we launched four regions at the same time. So you can imagine two massive companies, big legal departments and product and engineering. And when you’re landing physical hardware in another data center, not even just under the data center, another cloud’s data center that is owned, operated by them and it’s your hardware and your own data center people, it becomes a logistical challenge. On top of that, you have to do all of the integration from things like identity to all of the analytics aspects of it, to platform integration, the networking making it seamless, all those things. And you think about the service being live, it’s crazy that we can launch this thing in such a short amount of time.

And then the regional aspect of it really just becomes a matter of customer preference and opinion. Now, most customers if you think about the Oracle database, they want extreme high availability within the region. So that means we want to have multiple different sites in a single location, so multiple data centers. And then they want to have DR capability. So they’re not going to go into production until they have at least two regions in that geography. So that’s really what’s driving the regional expansion. We have so much demand pent-up, I would say that we have no choice but to work on all of those regions in parallel. To give you an example, if we don’t go into India, we are completely eliminating the entire financial services market in India as an example. So we can’t just go launch in Delhi, we got to go launch Delhi and Mumbai and we have to launch multiple sites within those. So as alluding to it’s stripping hardware everywhere, making sure that all of our network connectivity is there. And making sure that customers have the ability to start doing things quickly and then be able to deploy to any region that they like, including MAA architecture which is a certification that we do in Oracle to make sure that the architecture is certified by Oracle. So we do that as part of the region deployment.

Patrick Moorhead:
I’m sure there’s people wondering, and I’m going to ask you this question, what is it like… I was supplying Google with infrastructure before it was cool. I think they were using Pyrax and Velcroing parts in there, and now Google has the largest data estate of anybody out there. And not everything is standard, rack-size, cooling. It’s very optimized for the Google infrastructure. And here you are putting in an Exadata machine into this environment. Can you spend just a couple of minutes of the challenges of putting… And by the way, this was the thing that blew my mind, they literally used this emoji when you kick the brain, blew your mind, blown, putting these Exadata infrastructure into somebody else’s data center. What are some of the unique challenges like power, what goes? What gives?

Karan Batta:
Yeah, I think at a high level just to sort of level set, this is not just an Exadata machine. So what we’re actually putting into Google’s data centers is, I would say mini-OCI in some ways. It’s an extension of OCI, but it lives in another cloud. So it’s not as simplistic as saying Google is just another core provider for us. But it’s sort of like that where, let’s say Virginia as an example. This is very true for actually a lot of different cloud providers where you have a parent site and then over time as more customers come, you need to go and get more data center space and you start getting child sites. This just looks like another child site for us. So from an operational standpoint, we’re able to manage that capacity in a very cohesive way. Our Virginia site looks pretty much the same as the site that we put into the Google data center. So it’s essentially the minimum footprint required for us to actually deploy a fully capable OCI data center or OCI region within the Google site. From there, it’s all about how do we hook in to the right parts of the Google platform, whether it’s things like the right network connectivity to make sure that everything is encrypted and we have the lowest latency possible because again, these are customers that are running emergency systems or financial tracking systems or massive retail organizations.

So they want the best performance and that’s what they expect out of Exadata. So the app cannot be living like 200 miles apart. It’s got to be right next to it, and then there doesn’t have to be overhead on top of that. So we have to make sure that there’s no additional overhead on the virtualization, or there’s no multiple hops going back to the OCI data center and then back again. So there’s a lot of optimization on the network side. Of course, there’s the physical components which is the actual OCI footprint, is still managed by Oracle folks. They’re still Oracle people, but we have to make sure that we adhere to Google data center physical policies, their security checks, and making sure we have the right path to our cage. There’s so many different nuances and complexities. You’d be surprised some of the times data centers get delayed or regions get delayed, it’s because while the cable tray that we use is not the same cable tray another provider uses because it’s slightly bigger or smaller or the power whips are coming out from the bottom versus the top. It becomes these physical challenges that become much more nuanced. Everybody builds their region very differently to somebody else. Conceptually it’s sort of the same, the goals are the same but we’re executing slightly differently.

Patrick Moorhead:
No, I love it. That just puts the exclamation point on how quickly the two of you are working here. So it’s been a great discussion. I think the last set of questions that I want to ask you guys about is about real world customers. This isn’t a press release, this isn’t a, hey, we’re fixing to do something someday in the future. It was GA in September announced that Oracle Cloud world, I was there. Great event. And maybe, Andi, we’ll start with you. Can you talk about real life customers, how they’re using the service? What are you hearing from them?

Andi Gutmans:
I think first of all, one of the reasons why we’re going so fast is because we have a lot of demand. So we’re going to India, to Brazil, to Japan, to Canada. We’re extending our regions in Europe. So that pace. And I think also to Karan’s point, we have to be very creative to be able to move this fast. And it’s been just a really awesome partnership and being able to do this globally in parallel so quickly, that’s really a sign of the customer demand that we have. We have customers like Dun & Bradstreet who are looking to use this capability for their databases. We’re seeing both a lot of interest in Exadata database service, so customers who have Exadata on premises when we transition to the cloud with this partnership. We’re also seeing quite a bit of self-service on Autonomous database. So it’s been nice. We’re both seeing the very large lift and shift workloads. But we’re also seeing customers who have smaller workloads and they want to get started quickly, and they’re basically doing a pay as you go with Oracle. So I think we’re seeing just a very broad spectrum of customers across industries and across geographies. And I’d say the number one question we get is, can you go faster? Can you launch more regions faster and let us land these? So it’s been very exciting.

Patrick Moorhead:
Yeah. Karan, how about from your point of view, can you talk about a specific customer or trends you’re seeing, how they’re using it, maybe surprises or affirmations?

Karan Batta:
I think if customers surprised us, I think we wouldn’t be in this job, especially as the core vendor selling the database itself. What I can say it confirms our suspicions, which is there are existing customers that are modernizing their states on Google, whether they’re doing AI or other kinds of workloads. But they have a gigantic chunk of existing Oracle footprint on-premise, whether it’s an Exadata or whether it’s just bare metal IaaS or whatever they’re deploying with an Oracle database on top. What we’re finding is sort of in two separate industries, like financial services is a big one, and the second one being retail, like running retail systems and then financial services. Both of those sets of customers are doing two different things. One is they want to move of course their Oracle database estate to either Exa or Autonomous, but then they’re running our back office applications like PeopleSoft or EBS or Siebel or those kinds of things. Those are fairly static and they last migration like 10 to 12 months. They run there and then they’re good to go. And then the second part of it is their custom applications that they’re building on top of Exa or Autonomous, but then they want to hook that natively into things like Gemini or Vertex AI or other types of path services around Google. So we’re seeing those two different sort, I would say almost dialects in some way. Customers moving their stuff back office, very static sort of legacy, but they need a place to run it. And then kind of modernizing in the future things like Autonomous.

Patrick Moorhead:
No, this is great. Guys, it’s been a great conversation. I really appreciate you taking the time. And again, as analysts, I got to call the balls and strikes and sometimes it’s tough love. But kudos to both Oracle and Google for really giving enterprises what they want, which is choice. And the fact that the two of you came together in a way to enable enterprises to activate their Oracle data, and also applications and intersperse that in what looks like a very seamless way, was a lot of work. But congrats. Thank you from enterprises for being able to do it. Even 10 years ago I was saying, ‘Hey, one of the biggest impediments to cloud adoption is going to be this pesky thing called enterprise data that for most corporations is still on-prem.” And it’s challenges with activating it. But anyways, thanks guys for coming on the show, and I will be at both of your shows. I’ll be at OCW ’25, I’ll be at Google Cloud Next ’25. Hope to see both of you there.

Andi Gutmans:
Great.

Karan Batta:
Likewise.

Andi Gutmans:
Thanks for having us.

Patrick Moorhead:
Excellent. This is Pat Moorhead signing off for The Six Five, talking about one of my favorite topics, and that is multi-cloud capabilities. Boy, have we evolved in the last decade. It’s great to see, and it’s all about enterprise innovation, delivering outcomes in an effective and cost-effective way. Hit that subscribe button, tune in, tune into all of our Oracle and Google content out there. Take care.