.jpg)
AzureSUCCESS
AzureSUCCESS
The Azure Data Estate
Making sense of Azure's many data offerings can turn out to be a more than daunting task, but listening to Sr. Cloud Solutions Architect Kevin Howell should give you a good leg up. Along the way we also discuss the joys of welding, plus the merit's of Kevin's unusual nickname: Stitch.
ABOUT: As a Data & AI Cloud Solution Architect at Microsoft, Kevin focuses on evangelizing the business value of the Microsoft data platform and providing technical sales support on a wide array of data and analytics products. Kevin’s passion and expertise are targeted on researching and implementing best practices for customers in order to deliver optimal designs for scalable, enterprise solutions. While not toppling data challenges of epic proportions, Kevin enjoys building & creating all manner of things from 3D prints to wood and metal backyard structures with his wife and three kids.
LINKS: Azure Synapse, Azure Charts, Azure Architectures, Power BI Data Gallery, Kevin’s SQL Site
CREDITS: Louis Berman (Host); Kevin Howell (Guest); Dan Phillipson / PremiumBeat (Music); Anne Lamb (Intro/Outro); East Coast Studio (Editing)
MORE: visit https://azuresuccess.buzzsprout.com/ for additional episodes, plus transcripts, and more ways to listen to the show. As to your comments and suggestions, please feel free to email your host, Louis Berman, at lberman@microsoft.com
We're asked oftentimes by a customer to do different things, to meet the customer needs. Every customer has different requirements, different ways. They operate different areas, regions, some are global, some are local. So from the data estate, we make sure, or at least what I've seen is that our engineering teams, our product teams listen to the customers.
Intro:You're listening to Azure success, the podcast by and for Azure professionals. Listen in, and you'll be sure to speed your customers March into the cloud. And now your host, Louis Berman.
Louis S. Berman:Hello, and welcome to another episode of Azure success. The podcast for Azure professionals by Azure professionals, I'm really happy today to have, as my guest, my colleague, Kevin Howell, a Senior Cloud Solutions Architect based out here in the Northeast, actually outside of Philly and Paoli a little in lovely town that I mostly associate with my wife's favorite assistant. So like that town. So Kevin is here to talk about the Azure Data Estate, all the data tools, all the fun things you can do with data in the cloud. But before we talk about that, I want to talk about welding. Kevin say hi, and tell me something about welding.
Kevin Howell:Good morning, Louis. So tell you about welding. So basically I have a long history of welding, I welded vacuum trucks, I welded trailers. I welded several things, but I picked up welding back in college cause we had to take an elective and why not do some welding? It seemed like a fun thing to do. So why not? I went to Virginia tech, which is smack dab in the middle of cow fields where everybody has to know how to do welding for farm equipment or whatever else and picked it up and I've been doing it ever since.
Louis S. Berman:Great. And dear listener, do not think that our little byway to talk about welding has nothing to do with Azure because Azure of course is all about construction, building things, making things, making them happen. So here we have a man who knows a lot about building Azure. And so what is the Azure data state? What does it mean?
Kevin Howell:That's a great question. I mean, to start with, Azure Data Estate, means a lot of things to a lot of different people. So it's not just, I come from a historical traditional relational database SQL server background, which everybody knows that area, but the Azure Data Estate is so much more, it's really encompasses analytics and encompasses AI and machine learning. It encompasses the entire compute structure, so container services, but also encompasses IOT services. And one of the biggest and fastest growing areas is Cognitive Services. So everything from text analytics to visual recognition. So lots of exciting things within the data space, almost all businesses run on data. I usually refer to it as the backbone of businesses because it really, and you might refer to the infrastructure as the backbone, but really all that muscle around
Louis S. Berman:That backbone is data. So that's sure where I talk about the data estate to the customers I have. So let me ask you a question because I'm a quote unquote, Azure expert took, the course got the piece of paper, whatever, in my own case, I'm an application programming development sort of specialist. There's so much of it. How do you even know which pieces to grab and touch and what is there? I mean, it goes off more than a page, for instance, in a recent listing that I saw, how do you grab it? How do you touch it?
Kevin Howell:Yeah, I'm a recovering app developer as well.
Louis S. Berman:<Laughter>
Kevin Howell:That's a virus it's hard to shake, but the longest short of it is it never should taper off or have a steep cutoff between data and application. And I straddled that line for many years and understanding that it's a cooperation and coordination between the two and being able to talk to, Hey, there are certain things that data services should be used for. In fact, a lot of my discussions with customers have to do with abstracting work away from applications, such as security, role level security, such as data masking. Those are things that for the past several years, maybe even decades have been done in the application layer, filtering those types of things and they really should be centralized in a data layer or things that you can do with an application. Yes, there could be a lot of coding to maybe do some text analytics, but now you have a data service that can do those text analytics for you in line with your application. So there are boundaries, but as the Azure data estate grows, the ability to sort of centralize what belongs in a data resource becomes a little bit more clear so that the applications can really focus on what they're good at, which is processing the user requests and accessing the data sets as necessary.
Louis S. Berman:Wow. It still feels like a lot.
Kevin Howell:Yeah, yeah. It's a lot. And if you're a data professional, it's going to be more so that's what sometimes is scary about jumping into the data state is there's a lot to Wade through. And our job as sellers is to be able to quickly and clearly describe what the different data resources are that we have in Azure and how best to use them. The use cases, the architectures that our customers can then leverage to build the best solutions, really that as cloud solution architects, that's what our job is. Right. I don't know if you'd agree, but when I go into customer, first thing I say is, well, what are you trying to do? A phrase that I use commonly my customers get tired of hearing me say is number one, think big. And then we're going to start small and we're going to go fast. I sort of co-opted that from a lecture and ready probably
Louis S. Berman:Three or four years ago, a Microsoft conference, an internal conference, our biggest one.
Kevin Howell:Yes. And I sorta thought that was pretty brilliant because that sort of captures that think big, start small, go fast and really captures the essence of what as a CSA I want to do with every customer, whether I have 50 customers or 10 customers or one customer, it's the same idea. Start fast and fail fast. Right?
Louis S. Berman:Well, so let's go into a whole nother topic. We have the what, there's just this incredible amount of great goodness and an Azure data state sort of sense, but I'm a customer and I'm going to want to know why, why should I go with Azure? Why should I go with any of our offerings? And believe me, it's not Azure as a wholeness centrally. It could be Azure as a piece. Why, but before we do this, I want to interject and we'll get back to the question I want to ask. Why should we even yes, mr. Howell, how like how I believe, why we should listen to him. This is a man who self admitted. His nickname is stitch. Why is your nickname? And again, listener, I'm going off on a tangent, but trust me, this is relevant.
Kevin Howell:Number one, the operative is, was stitch not
Louis S. Berman:Excellent. Excellent, good job.
Kevin Howell:I did happen to have a few incidents as a child where I was a little bit accident prone. It's something that my son carries on to this day, but I happened to have more stitches and more trips to the ER than my entire household combined. And it had a large part to do with, I had a real pension for exploring everything that I could get me into trouble or put me in harm's way, because I'd like to sort of push boundaries if you will. So I would say that tapered off in my mid twenties, probably as most people, but I like to think I'm a little bit less accident prone today, more wise, but my wife might have other suggestions.
Louis S. Berman:Well, I find it impressive. And of course the stick to it, and this is a very important aspect in Azure. Azure doesn't happen overnight, right? And Azure, hopefully you won't hurt yourself nearly as much as Kevin has in his earlier years. He's reformed yours. I think you said. Yeah. The point is it's really important to get into Azure step by step. Say, why, why is it so important? What do you get? You're a customer. Why would you want to be an agile?
Kevin Howell:Yeah, I think that's a great question. We have a lot of customers that realize they have choices and I go back to why I'm at Microsoft. And I usually reference a large part of it has to do with Satya Nadella to be quite honest, I came to Microsoft about four years ago, shortly after Satya Nadella came to Microsoft. And really what I saw was a sea change in Microsoft itself. The whole, and what I saw was with the introduction of Linux to sequel and really the open source and the open arms to a lot of what was previously viewed as competitors. I think of SAP, think of Oracle. Think of, like I mentioned, Linux now sort of having open arms and having with our rivals and trying to figure out what's best for our customer. Putting customers first is really where I lived. I come from a very service oriented background and that's sort of, sort of how I've always operated. It just was in line with what I've used. So the question that you had, which was why, why Microsoft, because this perception, this view that we take on customers and to even more so, how we operate as a business is very much focused on how are we doing the right thing by our customer? How are we doing the right thing globally? You know, in these strange times, how are we navigating? You have strange and challenging. It seems like everybody's has different challenges today. And there's no shortage of serious and very introspective things happening in everybody's lives. Things that make us all think about what do we value what's important to us. So if I bring it back to the question of why Microsoft, I think we see a lot of these values layered into our offerings. So we don't just go to a customer to sell product. We go to a customer and say, what are your needs? What's happening with you? How are you feeling? And we make an effort. At least I make an effort with my customers. And I would say, this is how we build trust and how I feel that Microsoft and the account teams I work with really make a difference. And we differentiate from some of the other services is we actually have a personal investment in the customers that we work with. And I think that's valuable. That's great. So
Louis S. Berman:That's half of the question and I agree with you, by the way, the thing that whenever I pitch Azure and Microsoft who people I pitched overall value proposition, Microsoft really gets our customers, I believe. And we're working really hard for their success. Indeed. You and I both worked for a part of the organization. That's literally customer success unit. And indeed this podcast as your success is about figuring out as Azure professionals, how we can help our customers have success. So expand upon that, not just in general, but talk about the Azure data state, why the tools that we're offering, why the services, the processes, the way we're thinking about it, why that's useful.
Kevin Howell:Yeah, that's a great question. I think the crux of it is we're asked oftentimes by a customer to do different things, to meet the customer needs. Every customer has different requirements, different ways. They operate different areas, regions, some are global, some are local. So from the data state, we make sure, or at least what I've seen is that our engineering teams, our product teams listen to the customers. So if you look at power BI where all the engineering teams refer to the ideas website for looking at where to go and which services and features to focus on to some of our newer products, like sinaps, there are boards for customers, any customer to give feedback and input on what the engineering teams are doing. I've already mentioned Linux, but bringing Linux to SQL. Everybody thought everybody that I've worked with has always thought that was an impossibility. It never going to happen. The fact that we got there within a few short years of the executive change and leadership here is nothing short of amazing. So the fact that we can quickly agily shifts to what our customers are asking for and even more so bring to market products that are either gaps in the market or that our customers have specifically told us or needs that they have is just a different way to operate. And it's great to be able to be in a position where at Microsoft, we are constantly learning about new services that are bringing some great new features or feature availability to products that we currently have, or even new products altogether that fill a need. So do you have a favorite? There's this incredible press Oh, stuff coming along. So what has sort of tickled your fancy? I have to say timing is everything. Today is June 12th and I hate to date it, but today is June 12th of 2020. And as of today, but within the past few weeks at bill this year, there was several new announcements about sinaps, which to be quite honest with you, I've said this a few times, but I've been working in the data space for over 20 years. And this is by far the most exciting product and product announcement that I've seen. It quite honestly gives me chills when I talk about what it's doing, because if you've come from the data space from a relational data space, you lived in a transaction and trying to get after what's happening to my customers, what are the transactions that are going through? How do I get it at the analytics to better my business, right? That's always the goal. How do I get more return? How do I improve what I'm doing, looking for operational efficiency, but also at the same time business efficiency, how do I get after that? And as data gets bigger, how do I consume that? And actually use that to my benefit, to be prescriptive and predictive about what's going to happen in my business. So yes, we've had been able to do that through modern data architectures for some time, but what's really exciting is now a synapse. We're sort of putting that all together in one workspace. So one workspace, um, big Lord of the rings, fan one ring to bind them all one framework to do it all with synapse. So it's pretty amazing. I know whether you're dev ops or whether you're infra or security, doesn't matter what it is. If you look at what synapse brings to the table, the way that it really unionized is all of our data state stellar products that we have into one, you know, from data factory, to being able to use things like Python, Java, or C sharp, all within a notebook to get after your spark big data clusters, or just using standard to leverage SQL compute, and look at your visualizations through power BI. You're doing that all under one framework and putting it into one workspace is now going to be available. So that's something to get super excited about. And like I said, now that we can demo that and we can showcase end-to-end working with your data without having to move it around that's game changing.
Louis S. Berman:Wow. So let me take this from another angle, from my own personal experience. I like, I think you, and most people I've been talking about who have been an Azure space for awhile, I'm just going goggled. I'd over the number of ways that AI is infusing everything, right? And so let me give you Lewis's top, top, top, top, top, top, top AI improvement over the last three weeks. And I think thematically it's really, really important. I know for a fact, I didn't see any announcement about it, but I am a coder. I code in visual studio. I'm very familiar with it. I've done it for 20 plus years. And I've always had this painful thing where you try to use some symbol of some object and there's this thing called IntelliSense, which is supposed to help you find the thing, but you can't find the thing unless it is brought into scope, basically what we call creating a using. And that was a manual thing. And I cannot tell you how much grief over the years. It is literally a month of my time has been spent in aggregate manually[inaudible] and then just a couple of weeks ago. And in a way I didn't really notice suddenly some usings just magically started showing up without my having to find them type them do it and everything that is Microsoft AI infusing, not just the big processes and not just these enormous tools like synapse, which look at everything, but just a little bit of AI sprinkled on it. One of my colleagues on it, Ben Brown, who is a thought leader in dev ops, always talks about sprinkling the little dev ops on it. Well, I'd like to steal his phrase and say sprinkling a little AI on it. And I think we're seeing an incredible amount of AI at every level in Microsoft too. And so that's part of our value proposition, I think too. Would you agree?
Kevin Howell:Oh, absolutely. I think right behind sinaps one of the things that I do, we have a CSA and a day sessions, and one of the things that got me really excited at this pass and internal conference that we had in February was they had a BA knowledge mining hero in an hour session. And it's basically learning knowledge mining using AI machine learning and cognitive services and spinning up basically a full end to end chat bot over a relational model, relational data frame over Databricks into a cognitive service and do it all in a bout an hour, probably give it hour and a half, two hours. But the idea is you can spin these things up very now and there's so much focused on AI and there's so much focus on machine learning that really to put it in context, we're bringing new products to market at a rate that almost half of the data and AI sheet is cognitive services now. So well you looking at it in that frame that comes strictly from customer demand customers using things like HoloLens to look at their data. And if you're not familiar with Hollands or haven't seen the HoloLens demo, highly recommended HoloLens to what we can do on things like manufacturing floors or helping people understand how to train users on everything from mechanics to heart surgery. It's very interesting augmented reality, and we can do things in that space very quickly to work with different customers from almost every practice is it's just amazing what we're doing in this space.
Louis S. Berman:Great. So let's wrap this up with my final question. We've talked about the watch. What is the Azure data state? The why, why Microsoft is great for it, but we're sellers, we're Azure professionals where people who were either convincing people in our companies that Azure should be adopted or outside of our needs, that Azure should be adopted. How do you go about making the value proposition four, those people, right? The people who are not on board yet,
Kevin Howell:We have conversation as it really depends on the solution they're looking for. So I care to do a lot of discovery. I mean really what I would recommend to any seller. And I'm sure a reseller is probably doing a lot of this already is listen to the customer here. What they're trying to do here are the challenges that they're bringing up and speak to those directly. So that for most data conversations it has to do with the data challenge, usually it's a historical third party application or something that's tying them to the resource that they're using. And what I would suggest is don't just listen to what their pain point is, but try to listen to possible solutions they've tried. And also the story behind the story. What I mean by that is yes, they might have a pain point with this particular application, but what are they trying to get to? So don't stop at what's your pain point. Okay. Let me note that, but more of a, what are you trying to do? What would you like to see as again, the big picture? So when I tell customers to think big, I don't tell them to think in the framework of what they understand to be the constraints of our data estate today. Because what I tell them is, Hey, we have access to the engineering teams. We have access to the products and the roadmap, not where we might not be able to tell you everything that's coming down the line. We want to know from you, ideally, what is your best case scenario? And we want to hear that as Microsoft, we want to help you, you know, our job is to be there, to provide the solutions for you. So tell us, when you think about what your ideal solution would be, what does it look like? If you eliminate the middle pieces, the things that give you friction today, how can we help you eliminate those? And if we can get past that, if we can get to a common understanding, then Microsoft and any customer, we work with have a clear path forward. And whether we can get to a resolution on Microsoft or whether we can point them in another direction, whether it's a partner solution or a third party or something in the marketplace, there are numerous options and pathways we can go down. But the idea is let's work with customers to just give them options and work on the best solution for them.
Louis S. Berman:That sounds great. Well, this has been terrific. I've been speaking with Kevin Howell, senior class solutions architect at Microsoft, and certainly an expert at the Azure data state. And we hope you're going to have a really great weekend and there's going to be no reason to call yourself stitched on this particular weekend. And thank you so much for sharing info with us and we'll have contact details and stuff. If you want to reach out to Kevin directly in the show notes. So thank you very much. Thank you so much.
Outro:You've been listening to Azure success, the podcast by and for Azure professionals. You can visit our website, azure-success.com for show notes, helpful links and other episodes, but also to leave your questions, comments, and suggestions. Thank you for listening.