Clicky

Building Your Enterprise Cloud Adoption Strategy with HPE – Futurum Tech Podcast Interview Series
by Daniel Newman | May 18, 2020

On this episode of the Futurum Tech Podcast Interview Series, host Daniel Newman welcomes Alexey Gerasimov, Vice President of Global Cloud Delivery for HPE to discuss cloud consumption, the benefits of the cloud, and how to find the right mix of cloud for your business.

The Value of the Cloud in the Face of COVID-19

We have built a culture of face to face collaboration, but how we approach business and collaboration has changed in the last few months. Cloud has enabled us to be able to work from home. It has powered companies to keep going. We are coming to a point now where the real, true value of cloud can be realized especially those companies still looking to transform to be able to conduct business as effectively as possible.

Public Cloud Works… Most of the Time

Public cloud usage is through the roof right now. Originally, the whole concept of public cloud was centered on speed, agility, cost effectiveness and faster time to market — all of which was not available in on-prem environments. But that has started to change in recent years as on-prem, private cloud, and hybrid options have started to flood the market. Many companies now employ some sort of cloud mix to handle all of their needs. But the question remains, how do companies figure out the right mix of cloud options? Where should workloads go? This is what HPE strives to help companies figure out.

Enterprise Cloud Adoption Needs Strategy

Like any technology, just because you can migrate to the cloud doesn’t mean you should. Your enterprise cloud adoption needs a strategy. Shadow IT won’t work for the cloud.

HPE works with companies to figure out what exactly their needs are in order to recommend the right mix. It’s a multi-phase journey that allows companies to build in increments similar to a devops approach. Alexey broke down the basic steps of the process into strategic workshop, assessment, minimum viable cloud, final cloud solution.

The strategic workshop is the most important step. It lays the groundwork for the entire cloud strategy. Companies first need to answer a few basic questions to get to the strategy behind a cloud migration. Questions include why are we doing this? What problems are we trying to solve? And where can we gain the most business value?

Objective Recommendations

Once the strategic workshop and assessment are complete, HPE can pitch recommendations that are part of the Right Mix Advisor. The best thing about HPE is they don’t pitch one cloud over another for their own gain.

HPE partners with several cloud providers and pick the best cloud for whatever their customers requirements are. HPE determines what applications and workloads can go to the right cloud infrastructure based on decisions from tech components like security, governance, and finances among other things. They help customers find what they need and then can help the strategy shift as needed as well.

Process of Ongoing Engagement

What works today for a cloud strategy might not work next year or even in six months. Companies can spend unnecessarily more if they don’t pay attention when adding on or dealing with scale. Cloud is a living, breathing entity that HPE can help manage in a few different ways like cost control and managed compliance. Having the right structure in place from a cost approach is critical especially in situations like this where everyone is tightening their belts so to speak. Alexey noted that HPE understands this and helps companies get it right from the start.

COVID-19 and Cloud Transformations

With the outbreak of COVID-19 every company had to step up their game. If they had already started to transform, the process went faster. If they were already in the cloud, they likely needed more bandwidth or compute power. If companies hadn’t transformed at all, it became even more clear that they needed to. The spending must go on to keep business moving forward no matter what the situation is.

HPE stepped in to help even more. First, companies can defer payments for 90 days to get the technology and IT infrastructure set up to continue working efficiently. Second, if companies need cash upfront, HPE is offering a buy back program where they will buy back old outdated technology assets to help with the new IT infrastructure investment.

Furthermore, once IT leaders and companies realize what their cloud strategy looks like, HPE can work with them to figure out how that adoption happens. Whether it’s buying the technology back and managing it for them or moving technology to consumption models, HPE has a lot of different options available to help companies in whatever stage they are in.

If you’d like to learn more about how HPE can help you with your cloud strategy, please visit their website and be sure to listen to the full episode to hear more insights. While you’re at it make sure to hit subscribe so you never miss an episode.

You can watch the video here and be sure to subscribe to the Futurum YouTube channel while you’re there:

If you prefer audio, grab the podcast version here:

Transcript:

Daniel Newman: Welcome to the Futurum Tech podcast. I’m your host, Daniel Newman, Principal Analyst and founding partner at Futurum Research. And I’m excited to have you join us today on this Futurum Tech podcast interview series with HPE’s Alexey Gerasimov. Alexey, welcome to the show.

Alexey Gerasimov: Thank you, Daniel. Pleasure to be here. Really appreciate it.

Daniel Newman: Yeah. For everybody out there that is listening in, this is a live show that we also do on TV, so from time-to-time if you hear us talking at all about the fact that we are seeing each other, that is what you are hearing about. But I’m very excited about this show. We’re talking about the journey or the strategy for hybrid cloud and for overall cloud consumption by the enterprise. But before we get started, I want to first of all thank HPE, Alexey and your team for being our partner and sponsor for this particular podcast.

And I also just want to say it is the last part of April. It is the 29th of what feels like the 300 day long April where we have all been sheltered in place. I am here in Chicago. Alexey, remind me where you are.

Alexey Gerasimov: I’m just outside of Boston, Massachusetts.

Daniel Newman: Both of us in the … A couple of what you would say, the COVID-19 epicenters here in the United States. Not New York, but certainly in some of the hotbed. Both of us have been, like I said, not leaving our house much for some time, but fortunately tech as a whole has done pretty well in terms of being the enabler to help companies and people survive through these really difficult times. But I also wanted to point out that we are in the middle of this pandemic because sometimes when people listen to the podcast, you’re not necessarily hearing this in real time.

You might be listening to this and it’s June or it’s August. Hopefully maybe you’re in the future and you can tell us, are we free yet? But as you listen to this, just know we are in the middle of this so we will be talking a little bit more about COVID-19, it’s impact on business, it’s impact on enterprise, IT decision-making. And as we get further into what you’re doing, Alexey, with cloud, that may come in at least a little bit as a talking point throughout. I want to jump in with you here. This whole pandemic is everywhere. Our life and our work has been disrupted. Why do you think, Alexey, we can’t do our jobs as easily as maybe we’ve done them in the past?

Alexey Gerasimov: Yeah. Thank you again, Daniel, for having me. Really appreciate it. Back to your comment about June, if we’re still home in June or July, we’ve got bigger problems.

Daniel Newman: Oh, boy.

Alexey Gerasimov: Yeah. And I think the reality is that we have historically built a culture of a face-to-face communication and collaboration. It’s just ingrained in us that we have to go to the office to collaborate and do something, right? The way we approach business and the way we do our jobs changed significantly, right? And we’re experiencing this impact across pretty much everything. We’re disrupted from the way we deliver value, we’re disrupted in the way we communicate. And honestly, we’re impacted financially as well.

Daniel Newman: Yeah, I would say that that’s absolutely the case. It’s kind like a double-edged sword though, Alexey, because part of us, our life and work has been disrupted, but in industries like ours, tech has almost made it seamless. I used to travel 47 weeks a year, so every week I knew most of the flight crews on American Airlines when I was going to San Jose, when I was going to San Francisco, San Diego, in certain spots in Europe. And my point was, I traveled a lot, but when I wasn’t traveling, I worked at home.

And we’re seeing … A lot of companies work from home was either maybe a special privilege, like a flex thing that was offered from time-to-time, but companies were really pressed quickly to make that adaptation. If you were in that sort of knowledge work-type of role, companies just … Pretty much on Friday you went home, on Saturday you got a note and it said, basically, “Don’t come back. Get your laptop or we’ll send you a new machine and you’re going to log in and you’re going to do your work for home.” So it’s been somewhat of an adjustment for people, Alexey, But I would also say to some extent it’s been amazing how much the cloud and how much technology is essentially powered companies to just keep going. Of course, that’s not in every industry, but certainly in a number of industries.

Alexey Gerasimov: Right. No, I agree. If you think about it, just the benefits of the cloud and the capabilities in broad over the last few years, we know the various … Starting with IAS, the platform as a service too. Software as a service, you’re absolutely correct. We now have the ability to not depend or worry about our own infrastructure and to “outsource” a capability for you to do your job to someone else.

Technology specifically are more blessed and fortunate because we have adopted and embraced that approach for quite a while, but that doesn’t necessarily, to your point, extend to the rest of the IT [inaudible] business. If you’re a developer, you can write code and hopefully you’re adopting cloud-like approach, whether on a public cloud or on-prem anyway, so you can write code anywhere. But most of the folks in the general business operations in IT who are dependent on going to the office, I think they’re slightly behind the curve and they’re struggling with it more than others.

Daniel Newman: Oh, yeah. There’s certain jobs where it was impossible. There’s others that I think people need to be given awards. I have multiple daughters in high school. One’s in junior high, trying to go from classroom to e-learning overnight. I don’t wish that on anyone. I do think there’s potential for it. But here’s something I want to ask you about. So, people think about cloud, they tend to think about public cloud.

They tend to think about applications that can be consumed and that’s probably more as a consumer, because we’re thinking of the Netflix’s of the world and we’re thinking of the Facebooks and things that we are using in a very pure public cloud fashion. But that’s not really what it always looks like. So, when it comes to public cloud, is it a silver bullet? I mean, could companies have just gone public cloud and this would have saved and made everything perfect for their transitions or what’s your take on that?

Alexey Gerasimov: Right. No, that’s a great point. And I’m sure you read the same articles as I do. You hear the stats lately that they … With this whole COVID-19 work from home, that the usage of public cloud is sort of over the roof. And frankly, they’re probably true, that there is certainly significant uptake in the use of public cloud capabilities, whether it’s Azure, or AWS, or GCP, or even Alibaba, or all the other ones. But you have to be careful how you look at the numbers. And public cloud, if you think back to when Amazon first started, the whole concept for public cloud was agility, time to value, speed to market, cost savings, right?

Even though the cost saving might not be as relevant now, maybe more of a cost effectiveness. But public cloud provided you those capabilities and that was the allure of it, right? And it’s a cloud experience that was not present in the past on premises. That has changed significantly and the capabilities of on-prem environments are starting to match those of public cloud. Maybe not quite pound for pound or dollar for dollar, but they’re very close. So to your point, is public cloud the panacea of all evils? Well, no, but it absolutely can help you get the most value in the right circumstance.

Daniel Newman: Yeah, I think that’s … The big wait is when it comes to enterprise, it’s not really one cloud, right? As an analyst, we’ve done a lot of research on this. We’ve actually worked on some research with HPE. It’s going to be a series of different versions of cloud. Most people want the consumption capabilities. The company wants to be able to work in a single control plane, wants to be able to meter usage and understand how quickly it can scale, but the companies don’t necessarily or are not necessarily prepared to migrate all the workloads.

Perhaps there’s governance and compliance reasons, that public cloud is not an immediate option. This is something you specialize in. I want to hear about this more from you, Alexey, you and your team. But public cloud can’t address all of your needs in most cases. And as we see these accelerations of companies migrating to cloud, looking for that silver bullet, there’s some things they have to think about. And I’m sure in your experience you can probably give them some of the questions almost that they should be asking as they’re starting to migrate before they even think about procuring or deploying any workloads.

Alexey Gerasimov: Absolutely. Just to give a bit of a background on me. So, I came to HPE a couple of years ago via the acquisition of Cloud Technology Partners and we’re now part of CTP Practice, which is under the advisor and professional services organization and Pointnext. And as you know, HPE has a pretty far-reaching strategy that encompasses helping enterprises from edge to core to cloud, right? And cloud means literally everything. Hybrid cloud. There’s no hybrid without public, but at the same time, it’s not the other way around. There’s no public cloud only and no private. So the truth is, any enterprise will end up having some workloads in public cloud and probably multiple ones, like you said.

And now you hear all the terminology of multi-cloud, hybrid cloud, distributed cloud, all sorts of clouds. So the point is, you’ve got to have workloads everywhere. The reality is though, you shouldn’t go there willy-nilly and just put them without putting proper strategy behind it and without doing the rationalization of why you picked a specific cloud, or specific strategy, or specific workload. And that’s what’s been missing in I think in a lot of enterprise cloud adoption. We all know the historic sort of a shadow IT, let me put stuff there. That’s not going to work well for you and let you actually figure out why you’re doing this.

Daniel Newman: Yeah, I think companies are going to have to be able to analyze. And I think part of the beauty, Alexey, of cloud is the ability to measure the analytics, measure the economics. And so, I was reading offline some of your comments and I wanted to definitely discuss this with you. I mean, once you come to the realization that your strategy … If you’re an enterprise of any size, you’re probably going to go to some type of hybrid strategy. We hear statistics, somewhere around 20 to 25% of workloads are currently in the public cloud and then the rest of them are on-prem or in some sort of private cloud.

What we know is as much public cloud adoption as there’s been, a lot of cloud or a lot of workloads still remain on-prem. That means most companies are going hybrid. There’s a need, Alexey, I’m guessing to understand what that mix is and what that mix should be. Just because you can cloud, doesn’t mean you should cloud. And then, the economics of it, right? Because most businesses aren’t HPE. Meaning, they’re not in the business of IT. IT is an enabler of shipping more units, of building more products, of delivering more services. That’s what the IT is there to do. So that’s something, from what I understand in reading from you, is that’s what you’re really focused on is helping companies sort of understand those things. And that’s what you and your team are doing at HPE.

Alexey Gerasimov: Correct. Yeah. And one of the biggest things we help people get through is that famous … I’m sure you’ve probably read it and most of your listeners probably read it too, it Starts With Why, book, by Simon Sinek.

Daniel Newman: Love it.

Alexey Gerasimov: It’s a great book, yeah.

Daniel Newman: Oh, yeah.

Alexey Gerasimov: So, the question is, why do you want to do something? What are you trying to achieve? Because like you said, just because you can move it to the cloud, should it? The example I always use … People ask me, “Hey, can you move this application to the cloud?” And I tell them, “We go to the moon, we’ll go to Mars. Yeah, we can move a few workloads to public cloud. That’s not the issue. The question is, how much and how long?” So, what we help people figure out is the whole digital transformation strategy and journey. What are you trying to do?

Because cloud for the sake of cloud is not relevant. Cloud for the sake of how to improve your value delivery and how do you improve your IT to be a more effective business partner. That’s the right answer. So we’ll help them figure out, rationalize which applications should go. We help them figure out the economics, the right economics for the clouds. We help them figure out the security posture. And last but not least, I’m sure you hear it all the time, people are arguably the biggest obstacles to cloud or to anything. So we help them figure out how to get the people on the journey as well.

Daniel Newman: So, talk a little bit more, indulge me a little bit here. What does that process really look like?

Because I think a lot of companies are probably starting to read and hear about this more. You mentioned a few things, a little bit to digest there. You got culture of the organization. You’ve got existing technical resources, capabilities, debt. And when you talk about people, you’re not necessarily just talking about culture, you’re also talking about technical capacity and capability. For a company like yours that’s really working on both the public, the on-prem, the convergence of it all, what does that process for a customer that’s really trying to determine the right mix, trying to deploy it and understand a timetable, what does that look like?

Alexey Gerasimov: Right. And interestingly enough, HPE, like you mentioned, is a big company. I think we have 61 to 62,000 people, as of today. So we’ve gone through some of this transformation ourselves. We have offerings and we have our own services in public cloud. We have a whole bunch of our own IT that we’re on on-prem. So part of that realization of what’s the right mix for us, also applicable to us. What we normally do with customers as part of our advisory and professional services organization capabilities, we take them on this journey. We help them figure out and start with the strategic workshops. Really, why are we doing this? What problems are you trying to solve?What business benefits are you driving towards and what value you’re trying to get?

And we help to take them on a multi-phase journey from the workshops, to initial assessments, to building what we call the minimum viable cloud. And then they actually migrate on what’s necessary. And the whole premise here, we try to follow in the same sort of cloud-inspired agile way. Let’s not build a Taj Mahal right away. Let’s slowly build and increment and drive the value that you actually need. The proverbial, remove the waste from the process and speed up time to value. So yeah, it’s workshops, it’s initial enablement, quick builds, dev-ops approach and helping them get there faster.

Daniel Newman: And I imagine you’re also thinking about things like governance security, because what I kind of like and what I like about OEMs like HPE being involved in this particular phase is you don’t have necessarily an interest in any particular cloud. You are a partner, you work with Google, you work with AWS, you work with Azure, you obviously can support on-prem. You’re tying it all together and there’s a lot of value there. Because if you sell Cadillacs, everybody should drive a Cadillac. And since you don’t sell Cadillacs, in the sense of an AWS and Azure, Google exclusively, you’re able to sort of say, “Maybe these workloads could best be served here. Others in the other cloud,” right? There has to be a lot of importance to that because you’ve got things like security, you’ve got things like compliance, you’ve got things like governance, data sovereignty.

You’ve got certain clouds that do things better. Public clouds, that some are really great from a platform standpoint and have tremendous capabilities from a platform. Others maybe have strength in AI. We know that some have much greater capabilities and you might want to be placing workloads in different places. Now, if I understand right from what I’ve read, that’s really part of your whole process is kind of getting to the meat of that. Of saying, “We can look across the spectrum of everything that’s out there and we give you that consistency of IIS.” But you may be placing workloads in a few different places, depending on optimizing for what your business is trying to accomplish.

Alexey Gerasimov: Absolutely agree. And one of the best things that works for us is exactly that objectivity. AWS, who we’re a premier partner with, so is Azure and so is GCP. As you know, Microsoft and HPE have a long standing partnership that predates cloud. So, absolutely. When customers come to us, we don’t necessarily pitch them one cloud versus the other because somebody paid us more. Honestly, we might not have any preference for what cloud the customer picks, as long as it’s better for them. And potentially what we can recommend all three, right? All three big North American public clouds. So as part of our right mix, which I’m sure you’re very familiar with, will help people figure out specifically which applications and workloads should go to which public cloud based on a number of decisions from technology components, to business value, to security and governance, et cetera.

And then when we do give those recommendations, honestly, we can look them in the eye and tell them we picked them because we believe they’re best for you, as opposed to we’re pushing it because we have an incentive to push it towards your cloud. And interesting you said governance and security. One of the biggest things I see, well both actually though. Security is a huge topic of concern for all the CSOs, because everybody wants to make sure that whatever data, whatever applications they put on the cloud is protected. And then the governance, it’s closely tied to security as well as the cost. Everybody’s nervous, “Hey, what if I go to the cloud and I overspend?” And it could happen, but it doesn’t have to. You have to have the right governance in place. If you don’t know what’s going on, then you’re right, you’re in for a big surprise. So we advise and help our customers with governance and security all day long.

Daniel Newman: I was going to say, there’s a couple of different types. There’s the external governance. If you’re in a certain kind of business, compliance and governance. And then there’s the internal governance, where you’re really trying to control and manage the business. There’s been an interpretation by the market that cloud is cheaper. It is when you’re getting started. But many companies have found out with scale, with bursting, it’s kind of like your cellphone when you don’t realize you don’t have international calling and all of a sudden you get a bill after calling your grandma in Poland and talking for three hours.

That’s the same kind of thing that happens with cloud, is it starts off very economically efficient, but as it scales up, it doesn’t tend to have those alarms. And so, that’s something that your team is focused on. And that was what I was going to ask you about next was after it’s all said and done, after you’ve helped the companies deploy, there’s a process of an ongoing engagement, whether that’s in the format of true managed services or maybe more in a consulting and advisory standpoint. I imagine you don’t set it up and bale. Part of your plan is to continue to help the customer manage that governance, that cost control. Because even what you might have optimized today in a year could totally look different.

Alexey Gerasimov: Oh, I absolutely agree. And I just want to make a little funny. I didn’t realize you got a grandma in Poland too. No, but you’re right. You ended up spending unnecessarily more if you don’t pay attention. So one of the premises on how cloud started, “Hey, it’s much more cost effective than that premise IT.” And the truth is, it certainly used to be. And that’s because just on-premise IT used to be just completely inefficient, right? We would stand up seven to eight environments from test, to QA, to pre-prod, pre-pre-prod, then you name it. The reality is cloud could be cheaper, but you got to pay attention. It’s like when you go to the store and something’s on sale and you buy two items instead of one, you actually ended up paying more. But, “Hey, it was on sale. I got extra.”

So we’ve noticed the same tendency that people do now with the cloud because, “Hey look, this instance is only 10 cents an hour, let me grab three because for whatever reason I need them.” And then you forget to turn it off. So the bill definitely comes up pretty, pretty high. So one of the things that we do upfront, and to your point, it’s a whole journey. It’s getting to the cloud and staying in the cloud. So upfront we help people figure out what’s the right structure from all sort of perspective, from architectural perspective, from cost approach, from prepaying for reserved instances as an example, right?

But then afterwards, it’s a living, breathing entity. You can’t just walk away, you have to pay attention. So we have a couple of services, managed services that we offer to our customers if they want to. One is a cost control and one is continuous compliance. That’s to allow you to pay attention. Not necessarily turn it off, but to pay attention and be aware of what you’re spending. Just because you’re spending more, that might be a good spend. That’s okay. But what’s key is you have to know where the money goes and what you’re spending it on.

Daniel Newman: Yeah, that’s a really good point. And speaking of where the money goes and where do we spend it on, there’s a lot of talk about acceleration of digital transformation. But one thing we do know for sure is companies, with the exception of a few industries that have really benefited tremendously from what’s going on with COVID and the economic shutdowns around the world, are going to be in a difficult situation. If the companies were able to sustain, it doesn’t mean they’re in great shape. But as I suggested in the beginning of this show, Alexey, every one of these companies is going to be like, “Well, we need to up our game in technology.”

Whether that’s greater levels of security, data accessibility, machine learning and AI tools, automation that can be used, companies are going to realize if they were lagging in their transformation efforts, it is too risky to not transform. That means tech investments are going to be required. HPE had some pretty interesting announcements. I want you to talk about it a little bit. But it seems to me like HPE has recognized what I’m recognizing. They’ve recognized that companies are going to be in a more difficult situation, but the spend must go on. So talk to me about kind of what the approach is, you, your team, the company to enabling companies to make these investments to move to the right mix of cloud in the wake of everything going on with COVID-19.

Alexey Gerasimov: And to your point, even pre-COVID 19, HPE realized that people will figure out the right placement of their workloads and their data. It might be on-prem, it might be in the cloud. And if you decide to go to the cloud, well we go … We’ll call it the agility bubble. You have to invest upfront to save some money later, right?

But to invest up front, you got to find the money somewhere. So HPE historically offered a number of ways for people to actually get the money upfront. It could be everything from asset buybacks, to potentially switching to a different consumption pay model on-prem, right? To give you cash. But with COVID-19, I think we’ve done a great job and took it even further. The obvious ones, if you’re thinking about the big two buckets that people need to worry about, how do we get through this and then how do we grow and transform?

So that the, how we get through this bucket, in a way allow folks to defer payments up to 90 days in the current environments that they have, will allow folks to invest in their new IT infrastructure now and not pay till later with favorable terms throughout 2020. We’re allowing people … Well actually, we’re buying back from people their assets.

If they need cash up front, we can sell them more efficient software and hardware by buying back their old sort of outdated deprecated assets. And then frankly, we just let them take possession of our technology right now and try it and install it without actually paying for it upfront.

So that should help people in the immediate, pick it, three to six months sort of range to help move past the COVID-19 problems and into their initiatives. And then as you start thinking about the bigger transformation, back to my agility bubble. Once people realize which cloud is right for them or what their strategy is, we can certainly work with them to potentially reallocate how that adoption happens, how to minimize that agility bubble, whether it’s buy back their entire estate and run it for them. Maybe moving some to our GreenLake consumption model or various other mechanisms we have.

Daniel Newman: There are certainly a number of different approaches and that’s something I’ll leave for you to share with all the viewers that followup from this conversation. Alexey Gerasimov, thank you so much for taking the time to join me here on the Futurum Tech podcast interview series. You’ve been a great guest and clearly what you’re doing is making a difference out there.

Alexey Gerasimov: Thank you, Daniel. Really appreciate it.

Daniel Newman: So for everyone out there that wants to learn more, go ahead and check out the show notes from this podcast that I just finished up with Alexey here. I want to thank HPE again for being our partner in this particular episode of the Futurum Tech Podcast. I want to tell everybody, stay safe out there. Hit that subscribe button. We’d love to have you as part of the family for both these interviews that I’m doing, interviews that others on the Futurum team are doing, and of course our regular weekly podcast. But for now, I got to go. I want to thank everybody very much for joining me today on the Futurum Tech Podcast Interview Series. We’ll see you later. Bye-bye now.

This podcast is part of a special series focused around what leaders and companies are doing to help employees and customers deal with COVID-19. Be sure to subscribe so that you don’t miss out on amazing insights.

Disclaimer: The Futurum Tech Podcast is for information and entertainment purposes only. Over the course of this podcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we do not ask that you treat us as such. 

Image Credit: AiThority.com

 

Daniel Newman
Close Menu