Customer Discussion – SAP Cloud ERP Customers Prefer CDC with HPE GreenLake SAP

So welcome. This is a section four for the best kept secret. My name is Jim Lana with HPE, Kelly Smith from SAP and, and Randall Grogan is on the, on the line.

And, and this section is for Randall to talk about energy transfer, his company, as well as why he chose, um, this solution. And so, Randall, I'll, I'll just queue up, you know, what, what's on there. And then I, if, if you wanna just, uh, cover energy transfer real fast and then I'll, I'll just prompt it for the other questions.

Sure. So energy transfer, if I, if I were in the room there, I'd ask for a raise of hands of how many folks know energy transfer. Very seldom do I see.

Unless you're in this industry. Yes, you, Jim, don't count. No handle.

Very, very seldom do I see folks raise their hand. So, you know, we are, by most measures, the largest midstream pipeline company in North America. I think current, we're about 135,000 miles pipeline, predominantly natural grass, crude, NGLs, kind of in that order.

We're in the Fortune 50 range, plus or minus, about 90 billion. Um, we've got some large strategic export facilities, Marcus Hook, deland. Um, we have a large first purchasing business.

We purchase about 30 to 40 different 30, 40,000 transactions every month of crude, you know, from small mom and pop, uh, organizations. And we use that as supply for our pipelines. Uh, we're an MLP and, uh, and multi limited partnership.

And that model really, uh, promotes growth. So we do a lot of organic growth and we do a lot of acquisitions. Every year we do one or two acquisitions, but I Can, I, I, I'll interject one thing, Randall, like, if we talk about, this is a really appropriate conversation.

So who provides all of the power for these data centers and, and AI with all of this additional power we need? Oh, that's right. A lot of it's generated from natural gas, from energy transfers, pipelines.

We're actually assigning a lot of deals, a lot of them in Texas and Oklahoma, really around providing natural gas because a lot of, uh, a lot of the data centers are, you know, in and around Texas and they're all using natural gas generation in many cases to provide that. And we're signing deals to provide that natural gas and are long term deals that we're signing. Um, and, but I was gonna say we're, I'd be remiss without really noting that, you know, I'm talking about energy transfer, but, but, you know, Sunoco LP is a one that everybody does recognize.

Um, Sunoco is a fully owned subsidiary of, uh, of energy transfer. They trade on their own, so they don't show up in our stock. But, you know, they're in the $25 billion revenue range as well.

They just made an acquisition that's gonna double their size, so they'll be in the $50 million, $50 billion range, excuse me. Um, and, you know, making them kind of the largest, uh, um, you know, ref, you know, in, in their class. Um, so well, we get USA compression buyer standards relatively small at a billion.

They provide compression services to the industry. We're their largest customer as well. So all in all, i, our systems support not just energy transfer, but Sunoco, USA compression, SEC.

So, you know, we are, when we kind of finished consolidating this lattice acquisition to the system, we'll be about 140, $150 billion entity running on a single instance of SAP. So that's, that's energy transfer. And, uh, um, yeah, so, So go ahead if I Yep.

Thank you for the, for telling us about energy transfer. And, and so when you, when you, when you look at all of everything you've got going on in your business, why did you select RISE at the time or cloud ERP as it's newly renamed in the, in the CDC option? Yeah, so if you kind of look back about five or six years ago, we did a big application rationalization.

We took multiple instances of S-A-P-E-C-C and 50 other applications and, and basically consolidated in the one single instance of S four hana. So by energy industry standards, that was pretty early on, and that's five or six years ago. So of course, it's inner data center.

Um, we're running HP hardware and everything's running on that hardware in our data center right now. And it continues to run, but it's coming to end of life, which kind of gets us to where we're at right now. Um, you know, SAP has been asking about organizations moving to rise, and, you know, our hardware has a five, six year end of life time as time span.

So for us, if the conversation was gonna happen about moving to, um, you know, into a CDC model or for that matter into a hyperscaler model, it had to happen really in the 20, 24 time range. So from a timing perspective, that, that timing really worked for us. Um, and, you know, somebody mentioned risk.

Um, you know, we, because we are provided, we move about a third, if not a little bit more than that, of all the hydrocarbons in North America, um, we're considered critical infrastructure. So I know there's some cybersecurity folks in the audience here. You know, for us that's incredibly important.

Um, all of our commercial systems sit within our data center. Um, you know, I've got right angle and, and tarot and Allegro and all align and, and you know, a bunch, you know, synthesis and all these other very oil and gas sort of applications that feed probably 98, 90 9% of all the transactions that come into my system, the SAP system. So for us, latency is, is, is a concern.

'cause everything that happens and everything that really comes in our system is largely coming from these commercial systems. Our data center is a standalone data center that we own. Uh, we actually have third parties that co-locate in our data center.

Um, and you know, for us it's actually a bit of a profit center, probably a wash when you get all said and done, but it's also our backup both in Houston and Dallas for, for our pipeline control group. So we have to have a backup facility. So, you know, really, you know, for some organizations moving things out of a data center and into the cloud really is a cost savings for us.

The data center's still gonna be there, there, so it's just the, you know, as long as we have the additional floor space and that sort of thing, there's no opportunity cost. So for us, you know, the, there, there was no financial reason to move it out of, out of the data center and, and into the cloud. So, um, you know, by and large, you know, we, we feel that keeping the, keeping SAP and keeping the hardware in our data center provides us with lower latency.

We feel like we have a better opportunity to, to kind of manage the cybersecurity risk around that. 996, which is very, very similar to historically how we've been operating within our data centers. So, you know, for us, it, it just felt like we, if there was an issue, we would have a better opportunity to, to, to manage that issue to, to kind of, uh, um, you know, be the first in line to, to get it, to get it corrected.

Versus in hyperscaler, you know, you're one of many. So right or wrong, we feel like we have better control, um, by doing that within our own data center. Um, you know, part of our, you know, deal with having the conversations, and Kelly can attest to that, we, we were very insistent that our transition to this rise cloud ERP model needed to be cost neutral for us.

So we, you know, we went through a lot of back and forth in terms of what does the cash flow for this look like? And, you know, we, we felt like from a timing perspective, this work, and we felt like from a, you know, from a cashflow perspective, it was, it was neutral. And, you know, ultimately we've, we've invested in SAP, you know, s our s our ERP, and you know, we, we believe in it.

Um, you know, SAP is clearly making, um, making it financially, um, um, you know, imperative that folks move into this direction. Um, you know, both sticks and carrots. Um, and you know, one of the carrots is, is ai.

So, you know, part of the cost to achieve AI is a movement to this rise model. And, you know, we, we don't know exactly what AI is gonna bring for us, but you know, we, we believe it does have value. So I'll stop talking for any Question.

No, no, that's, that's great. And, Uh, how many racks of infrastructure are part of the implementation of, of the, the cloud, private cloud, ERP? You know, I believe there's three racks in our Houston center, two racks and our, uh, uh, business continuity center, Dallas.

And, and, and that supplies the disaster recovery for the SAB, um, solution. Yeah. Between Houston, Dallas, yes.

Okay. We have a four availability zone model now that is approved too. So, you know, some people are looking at doing the, you know, patching and rotating once, and then you stay there, and so you don't have to patch, you don't have to rotate, patch and rotate back.

So there, there are different options available in that. Um, but, but Randall, you know, it's important to know too, Randall is a GreenLake, an HBE GreenLake infrastructure as a service cloud purchaser from a direct contract standpoint. And also now because of the way that this GreenLake infrastructure as a service works under SAP, he's got that same GreenLake service, but through the SAP contract.

So it was kind of a logical extension of the way that energy transfer thinks of cloud being the right cloud for the right workload at the right time. And, and like most customers, they're using hyperscale public cloud and they're using private cloud. And, and this is, uh, just a great example of, of, of private, but, but Randall, let me ask you another question.

So what's most important for other customers to be aware of when they're considering SAP cloud ERP? Well, I mean, I, I Cloud ERPI mean, I think, I mean, obviously, um, it, it, it's kind of where you are in that, in that, in that journey, so to speak. So I mean, um, a lot of our, we, we buy energy industry standards are ahead of the curve.

Um, you know, most, most energy industries are probably back in the ECC, um, you know, really haven't even made that move to S four for us. You know, we're, we're already in, in that model. Um, I think, I think a lot of it depends on kind of, um, where, where you are in that journey, and obviously from, from a, uh, from an ERP standpoint, um, getting, uh, you know, getting to a version that is, that's, that's supportable by SAP, getting a version that gives you the value of, you know, of AI and that sort of thing is important if you're really considering the CDC model, it really boils down to, I think ultimately, are you large enough that you have the data center that can support it?

Are you, um, you know, are you, you know, do you have a complicated enough, uh, um, you know, environment that you need to consider, you know, the private cloud versus the public cloud? Um, I do think that it's gonna be a long time before organizations of our size and scale will be, you know, considering a public cloud, we've got too many customizations that we, we have to have. I I don't think there's any alternative right now.

Um, so anyway, uh, I'm not sure if that answers the question, Jim, but back over, Where's the data? I mean, do you have, um, governance overall, the SAP data for your private, um, cloud e rp? I mean, I, I, the question from my perspective, if we're talking about AI since someplace somebody is gonna wanna have access to ized data or something like that in order to train their AI solutions, are you, uh, signing away some aspects of data Security?

I hope not. The data's in our data center and, and that understand with us leadership, I understand that, but that doesn't mean it couldn't be transferred to No, it, it, it, it, it does not. Um, and contractually, it, it, you know, it will not.

I mean, obviously that's, that's a big concern for us. And, and you know, I think you'll see that, you know, as we kinda look at the growth of the CDC model, that a lot of the organizations that are heading that direction will be governmental, will be, you know, maybe, you know, in industries where you know that data and that privacy is very, very important to 'em. Yeah.

My, my concern would be if, if your, if your focus is to try to view it as, uh, an AI enabler mm-hmm. Somewhere, somehow that data needs to be used to train machine language, uh, neural networks and that sort of stuff. And, and is that training infrastructure gonna be part of this private edition cloud ERP?

Or is it is it's gonna be done someplace else? Kelly, I'm gonna, I'm gonna defer that to you. So the, The next No, Randall, I was gonna say the next section is kind of dedicated to the overall far reaching AI and how that kind of works in the world.

And, and so the SAP AI engine for their agentic AI is going to use your data and it's protected in its form the way that it uses the data. And so that is definitely SAP data, whether you would like to bring in your other data into that model or your other am AI model is up to you. I mean, I think SAP has a lot of data for an organization, and even if it represents 50% of the data in an organization, that's a lot of data.

So it has a good chance of being an important AI model. But there isn't a software provider out there today that does that, has, that doesn't have their own AI model. And so each company is gonna have to choose their own governance for how their AI works.

And if they're using sensitive data, whether it be, you know, formulas or particular services or whatever secrecy they have, they're probably gonna be thinking over overarching. And, and so that's the important thing is this is not restrictive, it's inclusive. And, and so when we talk about AI and you say you talk about, um, an LLM in the public cloud, you can elect to put what data, sustainability data, for example, you wanna bounce it against everybody else.

And so that's a good thing for saying, I'll release the information on sustainability out to the LLM and I'll get my benchmark against everybody else. But when it comes to energy transfer and where their pipelines are and things, they probably are very secretive and close, you know, close lipped on, on where that, where those are located, they're probably not gonna divulge that information on maps so that other competitors can say, oh, I'm gonna offer a lower cost option for routing natural gas. So I'm just saying that I think that it's important to give choices.

This world of modern cloud is hybrid, whether by accident or by design. And if you can't try to figure out what makes the most sense, where then you're probably gonna fail on the model there. It just, it's, it's, it's too complicated.

So I have a little diagram that shows technically how you would share information for AI through the integration center for SAP, and you can use your local data with it if you want to, or you can bring the data back from SAP to your local AI model if you want to use that more sensitive model. Uh, before we will go further. Uh, Randall, I have a question for you.

Yeah. What benefit do you see after this move for yourself? Um, I mean, ultimately, you know, I I, I think for us it is, it is less that we felt that we needed infrastructure or as a service.

I mean, I've got a very strong infrastructure team. I've got a strong networking team. We have a, we have, you know, a good data center.

So I, I'm that, that for us was not really where the value the business case, you know, lied. I mean, I think for us, it really boiled down to, um, this is the roadmap that SAP is heading toward. Um, you know, it's been made very clear that that, you know, new functionality is gonna be made available to, you know, those that are following this, this cloud ERP roadmap, um, AI being one example of that, um, that, you know, as you're purchasing new functionality and, you know, we do a lot of acquisitions, that it's gonna be much more costly to, you know, to do that outside of the kind of the cloud ERP model.

So, I mean, I think for us it's both a both, you know, new innovations and access to those as well as just the financial and the business case of, of not making that transition. So basically the biggest value you, one, at least that you're pointing out is that you didn't need to move to cloud. Yep.

Okay. I have a, a follow up. Hello.

Shala here is from, so to follow up on that, one thing I'm curious about, um, from a DevOps perspective slash you know, solution architect who just has not had to work with SAP mm-hmm. If you could ask, like, what was your overall experience with doing, you know, making the move? Like, did it feel like a, you know, like a one team situation?

Was there someone there to guide you? Like how was that? We're, we're still in flight, so just to let you everybody know.

So we, we, yeah. We are at the, we're at the process where, um, the hardware's now in our data center, um, you know, HPE is kind of, is in the next few weeks we'll be wrapping up, getting all of that, uh, ready to go, um, and handing it over to SAP to start their process of configuring the, uh, configuring the systems before it gets handed over to us in late June, July. So, I mean, I think so far the teams are all working well together.

We've got HPE, we have, you know, SAP and their delivery team all on calls at least weekly, although I've had meetings three times in the last two days this morning included. So I mean, that, that's, that's working well. Um, I mean, so this is a, it's a bit of a dance to make that transition.

So it's, it's not just, it's not just us, it's kind of us and HPE and SAP, and then you have to have a migration partner. Um, that's not something that as a client we can do. Um, you know, so we, we have a third party migration partner.

We, it could have been SAP or it could be any number of, uh, certified partners from SAP to do that. So at that point in time, we're gonna have four folks all working together to get there. All, I guess really three, because HPE is gonna be kind of done with their part of the job once that's transitioned to, uh, to SAP.

Thank you. I think a lot of it boils down to, since you have a lot of cooks in the kitchen, right? You need a strong governance model.

And that's something we've talked about from the beginning is, you know, in the pre-sales motion, who are the resources that are gonna see this from the day one, you know, post, um, and ensuring that many of those resources are still intact along the way. Because if we put a, for example, a new cloud architect into this right now, during this time, it would be such a challenge to get that person up to speed. So we try to keep it very consistent from that perspective and have levels of communication across our, our organization aware and very detailed of what's going on to be able to jump in, obviously if something does shake up, but that communication and structure has to be intact, um, from day one.

Nice. And then, um, one thing that I've been hearing, like throughout the presentation is, you know, that people see it as like something that's rigid. So like, what are kind of like ways where you're trying to, I guess, kind of make it seem, you know, like are you able to kind of like tip, you know, dip like a toe in without going all in?

Are there like sandbox or crawl environments so you can kind of get a feel for it? Sure. Before you just like, Hey, yeah, let's just go all in.

Yeah, we have, we have customers that, uh, look at, you know, putting a, a minimum box in place, so to speak. So we have a minimum quantity, uh, you know, s four user box that you can get cloud ERP with that, you know, minimal purchase. And then they use that to figure out, okay, so for example, you know, one of the customers in this room today is doing just that, you know, they're looking at saying, what are the a agentic AI use cases that are going to come with S four?

And so they have a box and they're starting to look at testing some of those, you know, use cases and opportunities, you know, dipping that toe in to say, okay, is the value there, is this beneficial for our business users, et cetera? Because, you know, for so long it SAP's audience was it, it, it, it, but when we're looking at the transformation and capitalizing on AI and ML and ensuring trying to get standard business processes, our ecosystem is not just IT anymore. The business has to be involved with some of this, you know, decision making.

So we have to be able to show them from a usage perspective, see it to believe it and, and how that would benefit their day to day. So, um, I think absolutely we have those opportunities to be flexible and offer, you know, the, the sandbox environments. Cool.

Yeah, there's lots of choices now for how to dip your toe in the water, whether it's non-production or whether you're going to, you know, just test out the operational model. So maybe like, like Randall's team had been providing their own database administration and, and, and basis support, which is the IT piece of, of SAP security and things like that. And, and so now he's got an operational model to say, SAP's in charge of that along with shared responsibility from his team.

And those are different changes. So people look at that and like, wow, that's a lot of change to absorb in my organization. What if they don't do the same job we did?

Mm-hmm. What if they do a really better job and now people aren't valued as much? You know?

I mean, there's all sorts of things going on behind people's, you know, what they're not saying. But, but those are all, you know, ways to test and, and you don't have to go all in. I I always say the same analogy as you did, JLA is, uh, dip your big toe in the water.

You don't have to jump into the water all at once. You can, you can figure out what works best for your organization when it's a big decision like this. And, and those flexible options are, I mean, as of right now, today for the first time, we have all of those options on the table.

I, I've got a, oh, sorry. It made me think of an, uh, another question. Like what are some of like maybe the common like type blockers or hesitation that maybe you are, are you all are seeing with like making the move?

Um, so I, I think that there's, when there's complexity in these large customer bases and ensuring that there's a champion on the customer side for sure. So a lot, you know, you can have, um, users for example, who are dealing with, you know, clicking in 20 different things to get to the final destination. And they're saying, well, how do we fix this?

How do we fix this? But it has to get to a level of an executive alignment, ensuring that from top down, like, change is going to happen and we're going to go through this evaluation to see where those value drivers are. Um, that is first and foremost very important because if you don't have that, we're all on a hamster wheel trying to figure out how to get there and get that communication and alignment across the organization to see where opportunity is.

Um, you know, ultimately everyone has a day job. Yeah. This is an addition to a day job, right?

So you need resources and you need, depending on the scale of your transformation, you need a lot of resources. So how do you identify those resources and get them into a model where you have GPOs, you know, process owners and looking at those ways of transformation. So challenges of resources comes into play there of, of the timing in which they can do that while they have their day job, you know, day-to-day job, can they get out, you know, can they focus on this holistically?

So that's another area of, of, of challenge. And then it boils down to obviously working through a lot of the change that's going to happen from an operational perspective, because now you're giving your keys to the kingdom to SAP. Mm-hmm.

And that from a CIO's perspective is takes a lot of conversation and a lot of feel good. Um, scary it to, to be able to do so. It takes time.

Yeah. And I think that that comes into a, a, an area of commitment of that time and making sure the best decisions making being made across, across the organization to give those keys over. So lots of people, lots of processes takes time.

And can everybody be there to be able to make this transformation? So that's why oftentimes we're seeing this, you know, kind of lift and shift Brownfield or ECC move because it allows them to take a certain portion of the resources, a certain portion of the risk, and move that breathe. Yeah.

And then start figuring out, okay, based on that rise methodology of the people and the tools, how do I prepare myself for transformation in a better way? So let me break down, because when Cali says the ECC move, ECC is what they're used to running for all these years, they're comfortable with the functionality of the user feature set. She says an ECC move, she means they're just gonna lift up the EECC and we're gonna put it onto a HANA database.

From a user perspective, nothing changes, but what is changing? The database is changing. So it's slightly different to support.

Now you've different care and feeding for the operating system and the patching, and that is now moving into the operational model, which SAP is now supporting database administration and basis. And so that's a change, but it's still dipping your big toe in the water. We haven't changed anything for the rest of the organization.

Anybody outside of it doesn't really know what's going on. Right? And so that's, that's part of this tailored move.

And, and that's why I think this, this option is so great because it gives people the ability to not have to move their data center if you're already all in hyperscale clouds. Stay there if you want to, the, the decision for moving to moving data centers is completely isolated. There's no difference in software functionality services, SLA, there is no difference to access to innovation.

None of that is any different. And that's the most important part of this is stop making data center migrations part of this decision. It has nothing to do with it.

You can make it at any time you want to, but you don't have to do it in the same, at the same time. Well, like, so what I'm hearing is first often what needs to happen, which a lot of times is what needs to happen is, you know, there has to be a cultural change within the Absolutely. Organization, so that way can like all trickle down.

Absolutely. Yeah. Absolutely.

Uh, just a question. I mean, what happens in the case of a ransomware attack to the customer? And, and where does, you know, S-A-P-H-P-E GreenLake, how, and versus the customer, how they, I mean, it's obviously gonna be a full force effect to try to re to respond to this, but how does this all play out?

I mean, to some extent it could, the data could be corrupted, uh, and Hana it could be corrupted, and the customer data, you know, once an effect should aid reaches the network and cracks through, you know, whatever it is, firewalls, things start to spread. So The, the basis is shared. So, so you have people know where the data is, they know how often a copy is taken, whether they're doing a snapshot, whether they're doing a DR copy, whether they're doing an offsite backup.

We, we know where the data is. You're going to, same thing you would do in internally, you would test for the extent of the migration, what's corrupted, what's not corrupted, how much RPO, um, how you minimize RPO, and then you start bringing things back. So it's, it's a shared model from an administrative standpoint, so that the first exercise is to figure out what's good, what's corrupted.

And the second part of it is what's the plan we're gonna put in, in action? Are we, are we, are we failing to Dr, are we gonna stay here and try to just restore what we have? And, and that really comes down to, uh, an extensive conversation between basis.

If it arises though, and it's a ransomware, it's attack SAP take action, SAP is authorized to take action immediately. And I I will say it's a shared responsibility in the, in the CDC model. I mean the, and you know, referencing the Colonial pipeline ransomware attack, which was a, uh, a, a midstream as well, that's by and large, why, part of the reason why, you know, we're very adamant we want to keep things in our data center.

We feel like, you know, my cybersecurity, uh, counterpart sits in the office right door next to me. And, uh, you know, we, we feel like a lot of the things that we've done around our data center and around, you know, monitoring that, um, we, we feel like we are right or wrong. We feel like we have more security within our own data center than we have in, in the cloud in, in the hyperscaler, I should say.

Right. Thank you. We are A little over time, so I'll just give, uh, Jim and Kelly a uh, a minute to wrap it up.

Sure. All you, sir. Thank you, Kelly.

Thank you, Randall. Thank you all the delegates and everybody that's tuned in for this. Uh, if you have more questions, uh, my email is, is in the presentation, you're, you're welcome to email me with questions.

The delegates can certainly field all of your questions and are looking forward to your interaction. And, um, we look forward to helping people tailor their SAP solutions their way, regardless of what your choice may be. So thank you very much.