Cloud Crunch
Cloud Crunch

Episode · 7 months ago

S2E07: CXO Considerations: Multi-Cloud - Is it Really Worth It?


The promise of multi-cloud suggests enterprises should be able to run their applications and workloads in whichever cloud environment makes the most sense from a cost, performance or functionality perspective. But the reality of the situation can be very different in practice, as enterprises grapple with how best to make technologies created by competing suppliers play nicely together. Contributor and Analyst to Forbes, CBS interactive, Information Today, Inc., and RTInsights, Joseph McKendrick, joins today’s episode to give his perspective on the value of a multi-cloud strategy.

...involve Solve, evolve, Welcome to cloudCrunch, the podcast for any large enterprise planning on moving to or isin the midst of moving to the cloud hosted by the cloud computing expertsfrom Second Watch, Ian will be chief architect Cloud Solutions and SkipBerry, executive director of Cloud Enablement. And now here are your hostsof Cloud Crunch. Welcome back to Cloud crunch. Today Iam joined by my co host, Skip Barry. Skip. Good afternoon. Good afternoon toyou and we have a special guest today. Joe McKendrick. Welcome, Joe. Thank you.Great to be here. I look forward Thio enlightening session and maybe I'llprovide someone like Man who knows, uh to introduce Joe here. He's acontributing analyst to Forbes, CBS Interactive Information Today Inc andArtie insights dot com. He's also an author, independent researcher, speakerexploring innovation, information technology, trends and markets. He's aregular contributor to Forbes and ZD Net as well. Much of his research workis a conjunction with Forbes insights and information today covering topicssuch as cloud computing, digital transformation, artificial intelligence,Big Data Analytics. He is a co author of the S O, a manifesto which outlinesthe values and guiding principles of service orientation in business. And It All right, well, I mean, Joe, that's quite a background that you have there.Quite impressive. Yeah. Today you know, we're going to talk about some of thepromise of multi cloud kind of strategies. It suggests thatenterprises should be able to run their applications and workloads and whatevercloud environment makes the most sense from a cost and performance off coursefunctionality, perspective at the moment in time. So, in theory, that'show multi cloud is supposed to work. But the reality of the situation couldbe a very different in practice. As we all know, a Z Enterprises grapple withhow to best make technologies created by competing suppliers play nicelytogether. I want to hear more about that, of course. So let's let's kind ofdive into this, Joe, you know, again, thanks for being on our show. And Ithink we're going to get some great insights idea today. Yeah. Hey, I'msuper excited around. This is well coming from my former background. Allall cloud on Prem to hear now, working for, uh, you know, cloud native Company.So, Joe Ah, good place to dive in. Give us the contrast. Yngve you, if youwould, from application developer Thio i t centric infrastructure view, if youcould, how that's evolved and where that actually stands today in Realm ofHybrid Club. Well, interestingly, I think from the perspective ofapplication developers, you know the move to cloud doesn't change a wholelot in terms of the structure of their jobs. And the structure of what theyneed to do is simply a for lack of a better way to put a change of venuerather than applications systems functionality. Residing in the samebuilding in the data center, it's been moved out. It's been shifted outward toAmazon Web services and Microsoft Azure, Google Cloud or whatever. It's beenmost optimal. Yeah, so if they're already doing some sort of agile orwhat have you, they wouldn't barely noticed any difference exactly. Exactly.There's that first move. You know that first migration, if you will, and oftenit's It's seen as, ah, cost saving measure. You know you're moving fromthe cap ex to the op X type of model. You know you're going to a monthlysubscription plan, but as things move along. That's where you begin to getsome, uh, Now, sometimes some surprises, uh, in other ways their eyes begin toopen up a little bit. Mawr there's there's more potential for differenttypes of applications. More robust applications, uh, new types ofapplications, new types of innovation when this happens further on down theroad as Cloud introduces its...

...flexibility, Um, there's some oftensome I don't know if you call nasty surprises or some unexpected surprisesin terms of cost. You know, you're going to a monthly subscription modeland everybody begins to log on and start using these services. And nextthing you know, you're getting bills. These humongous bills from your cloudprovider that match the cost of buying the equipment in the first place, right?We thought PM sprawl was bad, let alone. Now you have cloud spend sprawl. Yeah,yeah, there's some kind of law I can't think the exact name of the law. Youknow, it just this. When you build an expressway out into a rural area, youthink it's gonna help reduce traffic congestion. But that expressway createsshopping centers and all kinds of sprawl along that stretch, so thateventually becomes clogged anyway. Yeah, there's a name for it. I forget what itis. They are some kind of Murphy's law way. See that in Atlanta all the time.We expand the highways and they get more congested. So just slowing down.Exactly. Exactly. That's how it works with you know, you have. You know, thisthis great functionality with cloud. And everybody starts building alongthis route this cloud route that you've established and, uh, that becomesclogged. And, you know, the cost began to accelerate as well. So similar toeurban sprawl. We have cloud sprawls we've touched on and then managingmultiple clouds obviously creates a lot of complexity for the organization.What are some of the areas that can help that you've seen out there as faras the management and even looking at operating systems and then differentexpertise that in industry or an enterprise needs to cultivate thioprevent these types of things? Well, that Z, that's a tough question. Theanswer, because, you know, the cloud providers themselves provide somepretty decent tooling to help monitor and manage what you're doing with theirservices. There's, uh, there's a lot of great monitoring and management tools.But when we didn't talk about multi cloud or hybrid cloud, even you havethese different environments and you're attempting to manage and monitorvarious different types of environments and that becomes, ah, challenge. Itbecomes a real challenge to ah so I departments and, uh, as well, I thinkyou see, the failing of adoption happened. There is. It's the If themindset doesn't evolve right, the technology evolves and the platformsevolve. And then if the, you know human aspect doesn't come along, This iswhere we see a lot of folks failing, right? Exactly. Exactly. And and we sawit in the on premises. I t world as well. You know, you had a point whereyou had many different operating systems. For example, in environmentyou had, you know, I'm gonna date myself here. But if you look back inthe nineties, you know you had eunuchs. You have these unique server farms. Youhad windows come into the picture, you saw mainframes. And these all requiredifferent forms of expertise. There's a different tools to monitor thesedifferent environments. There is the challenge of attempting to integrate,be able to move data between these environments. And to some extent we'reseeing that with these various cloud environments that are on the picture aswell. You know where the technology is? A little more integrate herbal, shallwe say in irreparable. But still, those challenges exist. The elusive. Yeah,the elusive single pane of glass Still Yeah, the elusive glass. And you have,you know, you have these, Uh, you have these environments, you have the, youknow, so called infrastructure. As a service environments, you have AWS andMicrosoft Azure as the classic examples. And basically, you know, that's the I'mgonna say bare metal. It's It's probably not the right term for thissituation, but I'm gonna call a bare metal. You. You have this bare metalserver. The server farms the cloud you can build upon, but you still need theexpertise to build upon those and be able to integrate those. Then you havethese higher level services. You know salesforce dot com, for example, hasforced and they also offer, uh, they're offering. They're getting into theplatform as a service environment as...

...well. Next week, there's anotherexample. You know, they're offering ways to build upon their environment.So everybody's kind of getting into this game giving developers these toolsto build within the specific domains. But you still need to be able to bringthat all together. Yeah, what's in the complexity of the world where we standtoday? Go in 2020 and 2021. What's a good way? Help enterprises as they'reevolving, really identify the correct cloud platform for them. Obviously,nothing is 100% accurate or ah, 100% given. But what are some practical waysthat they should be thinking to get to that? Yeah, correct cloud platform. Youknow, that's that's a tough one. Sometimes you have companies whereevery department sees a different cloud platform as its optimal platforms. Youknow, the marketing department, maybe see Salesforce and they want to buildmore upon salesforce. And what's going on there? Three. I t department may beusing a W s. You know, other folks may be using azure, and, you know, I'm notsure if it's really a good idea to go with one single provider, one singleplatform. I think it's actually probably a a better idea to have thismix of environments thes thes competing or these different environments toprovide more choices. I think what's happening now and what is reallyworking to the advantage of companies is this choice. There's a lot ofchoices out there, and choice is a great thing. It really is a great thing.You don't wanna be relying on a single cloud vendor. You know, you know, AWSmay have ah, credible array of services available and incredible tools. That'sgreat, but we need those choices. We need that competition in the market,those offerings to move us forward and and to assure that we have a way tomove applications where they're needed. You know, if we're not satisfied withthe cost structure or the services being delivered by one cloud provider,we need to have that option. The shift to another cloud provider, I think,choices great, and and that's what we have now. And I think you know, it'sgreat if it can stay in that state. So multi cloud and hybrid really is a goodstrategy going into 2021. Still, it really is. It really is, you know, andit starts with, you know, backup in recovery. For example, businesscontinuity I've always said, you know, the ideal model would be if you'reusing the cloud for the bulk of the operations, it's great to be able toback up and restore from an on premises environment. And if you're focusing onpremises, used the cloud as your backup and restore it, it's great to be ableto use both. We move into more business centric applications, artificialintelligence where you need that capacity. You know, it'd be great to beable to leverage the capacity. The resource is off the best providers outthere, you know, AWS provides Ah, huge storage capacity, for example, thestore, all the data that big data you need for a I But there's other services.Uh, you know, Salesforce's Einstein that that provide were targeted ai typeof functionality. Um, you know, you need that variety. You need thatability to choose. And, you know, I rued the day when you know we Seymourconsolidation. You know, hopefully it won't go that way. Hopefully working tocontinue to see a dynamic marketplace with a lot of choice. Yeah, and I thinkthat kind of add on to that trend to I think we've seen more and moreannouncements of one cloud provider being able to ingest data out ofanother cloud provider for a big data platform. And it seems like there's alot Mawr interoperability taking place than there was before. So to somedegree, you know, we see a lot of people saying our analysis paralysis,right? We don't know which Claude to go Thio Now I think you can kind of hedgeda little bit and say, Well, if I end up in this storage mechanism from thiscloud provider, this other cloud provider could still consume it. So Ithink it is taking a lot of that risk out of there as well. Decision,actually, and I think it's noteworthy was, well, that just about everysoftware application provider out there... coming from the cloud. I mean, Idon't think anybody buys new applications, uh, for on premiseanymore, you know, on a CD or, ah, disco, whatever. It just wasn'thappening where everybody is coming from the cloud. So, you know, if youuse ah 100 applications in your enterprise, you may be using 100different sources, a cloud based sources for your functionality, andthat's okay, you know that that that works. So So we're sitting in kind of ahybrid, multi cloud world, and, you know, we talked about analysis,paralysis, and that's definitely a real thing. And so anybody going throughthat out there, that just that's normal just to understand that, But let's talkabout some of the ways to identify the correct cloud platform for somespecific applications and workloads. You know, what are some of the kind ofdecision matrix that take place and and kind of What is that mechanism thatsomebody can kind of go through to understand where the right landingplace would be for those applications? I guess the defense where you're comingfrom, you know what kind of environment you're you're moving out of. If you'vebeen a primarily an on premises environment with a large developmentteam in house, you need to look at again that that that kind of a baremetal type of offering that I A s type of offering where you know you cancontinue your development work. You know, you probably a lot of proprietaryapplications or custom built applications. You need to move forwardand you need that you just need that foundation that I s type of foundationthat AWS or azure even Google Cloud provide to build on top of you know, it,it's called I passed environment, you know, uh, platform as a service whereyou have the database and the development tools and everything on onthe cloud. And, uh so for that for that kind of environment where you have adevelopment centric environment, you know, you need to look at thesefoundational type of environments, these foundational type of services. Ifyou're perhaps a small to medium sized business and you know you don't have ahuge I t. Department, um, you need to look at those types of services thosetypes of providers that offer more back end service, more service in the backend. You know, such a salesforce, which provides a lot of that from the backend. You know it again. It depends on where you're coming from, where yourcompany is coming from. And, you know, it should be said that a lot of thesecompanies you have a lot of these companies with huge developmentdepartment. They produce a lot of software, you know, the automakers forexample, are becoming software companies. And, you know, you've heardit said before that every company is now becoming a software company, and Ithink to a degree in the process, every company is also becoming a cloudprovider in its own right. You know, you have a financial service company,for example, that that puts APS out there and the APS to ride servicesessentially there. You know, they're probably leveraging back in cloudservices, but they're providing a service, a SAS or ah, cloud basedservice to their customer base, You know, through APS or uh, throughthrough mobile Web or or whatever. You know, it's, you know, they don't call acloud, you know, you make, you know, you have may have insurance providerthat's providing these interfaces to its partners, you know, so that theycould see, uh, what the rate schedules are and things like that. But they'renot gonna call the cloud service, you know, they're just za partner networkor, uh, in exchange or something of that, right? Yeah, I think you'll seeto your point, it's funny. We were just meeting with our CEO a couple weeksback and talking about EJ, you know, the more core and edge stuff and thatthat will drive. I think it'll be symbiotic of one another, or one willdrive the other vice versa, you know, But that that we'll see a lot of thatplay out as companies become mawr shared ownership of data. You know,they're not in the business of data collection. They don't want the risk.They wanna move it somewhere else. But yet they want the data. All that kindof stuff is Well, so yeah, well said, Yeah, and the internet of things isreally throwing. This whole thing opens. Well, we have all the data coming infrom your partners and good point there.

Who owns the data? You know, you know,if an auto company has a sensor, if if an insurance company rather has sensorsthat are planted within autos, who owns that data is the driver is theinsurance company. Is that the automaker? There's a lot about things Ineed to be hashed out with. That with your compliance way talk a lot aboutcompliance around here. It's only going to get more complicated as thisindustry of all that is for sure. Yeah. Yeah, And I saw formula once. Uh, I wason some online somewhere where you calculated the worth of your data. Likeyour personal data. How much is that worth on the market? I think mine cameto, like $8.25 or something. Z, uh, low risk, lower. Uh, it's interesting. It'sinteresting times in that regard. How if we take a step back and we look atthe cloud providers today AWS Azure, G C P. What are some highlight benefitsfrom your perspective that you've seen out there you've written about? You'veprovided some commentary? Um, yeah, benefits of each cloud provider arethere, you know, benefits of one over another, not tow, not toe pit, oneagainst the other. But when you see some quick winds, especially in theenterprise, why one would stand out as opposed to another one? Well, um, youknow, a Ws really has had a lot of the attention and a lot of the pioneeringwork. I guess you could say in the cloud space has been coming out of AWS.They've been in this space since, uh, I'm gonna say 2000 and 4. 2005. Theykind of pioneered the whole concept of moving out of the data center andrelying on servers and virtual service, if you will, from from another sourcefrom the cloud. And they've been really taking the lead with with pioneeringdifferent concepts, they've you know lately they've moved to ah, conceptcalled outpost, where, you know, if you if you need late and see if you're notclose enough to an AWS data center, you need that late and see you can amplifythe signal, if you will, you know they'll bring in a little not a little,but they'll bring in a rack of service for you and the service act as, ah,local cloud that accesses, uh, also AWS has the pioneering. The innovators edgeMicrosoft Azure is an advantage because everybody has Windows Windows Server onside. They have ah, sequel server databases. They have, ah, Microsoftcentric development environments. So Aziz is easy to plug into theseenvironments. You know, there's, ah, the Windows 3 65 Microsoft 3 65services as well. That has that include word and the productivity applications.Which kind of tying the sure, but not quite. But you know the very similartype of concept you know, you have these. Everybody in the world is justabout every company in the world for the past 2030 years has been aMicrosoft centric environment. To some degree. Ondas, you're kind of floatinto that so that you have that advantage there. G c p. You know Google,Of course, as with, Amazon is a real innovator, and, uh, they have the costadvantage. Typically, Google services have been, uh, free or close to freefor many applications and many uses. And, uh, you know, they have. They havethat cost advantage as well as being highly innovative type of environment.So among those three, you know, it's, uh, some some interesting choices. Sure.Thank you for that. I wanna challenge question for you. Not not to be toohard, but so does the S o. A manifesto still stand across. Doesn't matterwhich cloud you are, or that's a great question. USO, a service orientedarchitecture has been around for some time. It z people don't talk about itas much. It's kind of just kind of...

...blended into the woodwork, blended intothe foundation of things We formulated the S o a manifesto with Thomas Earltook the lead. Thomas Earl of our I'm gonna mangle the pronunciation are keyto, uh, education. He's based on Vancouver. He's been he's been a lot ofbooks and he pioneered the the idea of constructing and manifest very similarto the agile manifesto. You see a lot of similarities between the two. We metin Rotterdam back in 2009 and came up with these concepts. We had just aboutevery major vendor a swell, some problem. The analyst involved with theformulation of the principles. And I think, for the most part, theprinciples still saying justice, the agile manifesto, those principles stillstand And basically what the S O a manifesto says is, you know, don'tbuild your i t as a monolith. You know the I T. Needs to be responsive to thebusiness. It needs to be more granular. You need thio build as the business youneed to grow your i t with the business and stay close to the business and stayfocused on what the business needs. I think yes, so manifesto is, uh,foundation of cloud. It's basically cloud is s o A commercialized e wouldagree. I feel like we get lost sometimes. And we let the cool featuresand stuff drive. You know, our decisions where we need to go back andlet the business drive. What? We make the decisions as opposed to the featureset driving, uh, you know, better decision or whatever. You know that,but But I like that s a manifesto. Certainly. Uh, still stands the test oftime here in the cloud world. Yeah, yeah, yeah, that's great. So, Joe,going into, you know, we're going into 2021 at this point. And you obviously,you're out there doing a lot of research, talking to a lot of people.What are some of the trends that you're seeing there gonna either continuedevelop or anything else with any predictions that you like to share withus today? Okay, well, for those listening, we did this. We did thissession here at the end of 2020 late December 2020. So we're wrapping up2020. And what a what a year It's been a year it's been And, uh, for the i tindustry, you know, it's been actually ah, very. It's been a very forwardlooking year for the I T industry. You know, we talked a lot about digitaltransformation. We've been talking about digital transformation for adecade now, and it's always been something that's been a little bitfuturistic. You know, companies still have a lot of manual processes.Companies still have a lot of legacy infrastructure. Still a lot of workthey need to do in terms of corporate culture and adapting to the newenvironment. All of a sudden, in the spring of 2020 like that, it's like,You know, if you're not digital, if you're not residing on the cloud,you're not gonna be able to keep up with things. You're not gonna be ableto survive the impact of Kobe, which is basically every company, every employee,every knowledge worker. I should say I need to clarify that every knowledgeworker that's in an office was scattered among the diaspora. Thiscorporate diaspora, if you will across their homes worldwide. It's a globalpandemic, of course, and it's a worldwide phenomena. This this reallywas the ultimate test for the cloud. It was the ultimate test for digitalcapabilities. The ultimate test for the cloud and and the fact that at thispoint it's fortunate that a lot of companies were running on the cloud orrelying on cloud services, which were uninterrupted. No, Kobe could notinterrupt the performance. The services delivered by the cloud. You know, therewas capacity issues, of course, with local networks around neighborhoods.You know, where people are working out their houses and so forth Glitches likethat. But nothing major really...

...interfered with the performance ofbusiness. Um, the cloud being on the cloud really pulled a lot of businessesthrough this crisis. It's been really interesting to watch, you know, interms of disaster recovery, business continuity, working virtually. You know,if Kobe that struck 10 years ago, it would have been entirely differentstory. You know, we would economy would have been in a lot of trouble. Um, and we're gonna see that goingforward looking, looking forward in the 2021 you know? You know, hopefully, thethings they're going to clear up and this nasty viruses gonna become historysooner than better sooner than later. But we're going to continue to seevirtual work. We're gonna really be accelerate continue to accelerate inthe cloud. We're gonna continue Thio, move forward with digital and I've beencalling it You may have seen some of my posts. I've been calling thisphenomenon the contact lists organization. You know, we talked aboutcontact lists, commerce and contact list shopping, and that's been therequirement that's been needed. On Bet's we're seeing now is a contactlist organization where you can work virtually you can tap into all theseresource is across the globe. You know, Cloud based resource is people.Resource is and doesn't matter where people reside anymore. It Zaveri globalphenomenon and it's been it's been interesting to watch, and we're goingto see that develop. Yeah, and just a little footnote to that when we talkabout virtual work, full time virtual work. I don't think it's a great thingfor anybody under 35. E. Don't think anything for anybody. Starting thecareer we need. We need to have people be able to go into offices and interactwith each other and develop mentor relationships and things of that sort.But otherwise it za great thing. Virtual work is a great thing, you know,you just need to have that hybrid mix. Plus, if you're under 35 you might havesome toddlers running around as well. E O E o. Well, very good, Joe. Really.Thank you for your time today. Great insights that you have provided to ouraudience. We really appreciate that. Skip. Always good to see you and talkto you as usual. Thank you. Thank you. Thank you. Thank you, Skip. Thank you.And that's been great to be here with you. Absolutely. Thank you, everybody,for listening. And, as always, for comments, suggestions and ideas, please.Email is that cloud crunch at second watch dot com? You've been listening to Cloud Crunchwith Ian Willoughby and Skip Very. For more information, check out the blogged.Second watch dot com slash company slash block or reach out to secondwatch on Twitter.

In-Stream Audio Search


Search across all episodes within this podcast

Episodes (30)