Cloud Crunch
Cloud Crunch

Episode 0 · 1 year ago

S1E07: VMware Cloud on AWS & Virtual Desktops

ABOUT THIS EPISODE

We discuss VMware Cloud on AWS as a data center extension, for next-generation applications and Disaster Recovery, and virtual desktop solution, Horizon 7. VMware Cloud Partner Solution Architect, Frances Wong, joins us to get technical.

Involved solve evolved, welcome tocloud crunch, the podcast for any large enterprise planning on moving to, or isin the midst of, moving to the cloud posted by the cloud computing expertsfrom Second Watch: JESF, Aden, cofounder and Executive Vice Presidentof marketing and business development and Willghby chief architect, cloudsolutions and skipberry executive director of cloud eneblment and nowhere are your host of cloud crunch: Hey good after listeners, thanks forjoining us on cloud crunch, Jeff, Aven, cofounder, Second Watch and I'm joinedas always with Ian Willoughby and skip burry of Second Watch. Today we havespecial guests, Francis Wong, from themwhere with us to talk about Vm WorkCloud on aws Francis has worked at DM were for a number of years here, threeyears with the mwhere and currently is a cloud partner solution. ARCHITEC hasa deep it background, going back to her time at Texas instruments. Franciswelcome to the show- and why do you tell us a little bit about yourself andyour current role? Sure thanks thanks for having me that's right, I'm Francis Walg, I'mcurrently a cloud partner solution architect with VM. Were I've been inthis rule for about eight months and prior to this I was actually whattmmrfort refers to as a corp sc for or Stillicon Vallet for about three years.So some of my clients include really well known names, because it's out herein the big area, twitter arbanb lift. I did all the security companies fallsfor net checkpoint fruth point. I did a lot of semiconductors Ka like ten fourpantronics things like that, so about Li experience with customers with theirvm were infrastructure and in some of the entire BM or portfolio, and today Iam not direct customer facing, but I'm facing partners like Second Watch inhelping you with the vm where on a WS solution. So we want to help ourpartners not only learn about the solution and position it, but to makesure that you're successful in deploying the vmr cloud on aw solution right and today, it's friend center,all of us on the podcast and many of you. If not all of you are workingremote today as the world is fighted against covid nineteen. So it's kind ofrelevant that we're talking about VMC or Dmmore cloud on Aus, as this is agreat solution in a crisis like this, an a number of use cases where yousuddenly have to leave your on previce work environment that need access tothe tools, as well as the application of work loads. Many of the usecaseswill talk about today. You know is migration. You know even teyseterevacuation, your datacenter extensions for scalability, so next, tenapplication, disasetor recovery, remote workforce, obviously and many otherswell, Francis thanks for the over. Let's justjump in and just have you describe for listeners out there, whith the offeringdim, where you know cloud DMC on aws, provides the in customer sure sure so, Vm orcloud on aws just to break down that name a little bit when we say BM, morecloud, we're really referring to the VM or STAP. So the VM, whereinfrastructure stack the course tack is often also referd to as a BMWORK cloudfoundation and encompasses a three in frustructure software pieces. That vmwere is known for the first and...

...foremost being bpear, so virtualizationof compute and then there's NSX virtualization ofthe network and then there's vstand virtualization of storage, and thosethree is are the three main components that make up bea more cloud foundation.Now benmore cloud is an instance of v Cloud Foundation being executed on aWS's bare metal hardware, and we began about three years ago in thepartnership with a WS. Now this was not a small undertaking. It took hundredsof engineers from vm, where, in hundreds of engineers, from aws morethan twenty four months to bring that to market, so it took a long time andwe were a WS's first partner in delivering on metal as a serviceoffring. So the engineering effort I wanted to highlight, because theservice is almost turmty, and that is the beauty of all that engineering.When you turn it on you go into a web interface, you put it credit cart andyou sign up for a vm whore account and you can press a button and bring up theentire vm r stack in an aws availability zone. In about ninetyminutes, and that in itself is a hugeengineering feat in my opinion and when you say Engineerin an it's, not justthe m where engineers this was a joinet effort with as engineers and DM wereengineers and, obviously a very strategic Fr Ein if you've watched someof the media outlets with with your CEO as well as Benderia, ved and heard forAndy Jazzy right. This is this was a fairly large significant joint effort it. It is a big joint effort and we'rebreaking a new ground in that, when you have you know, Traditionallin vm MarasBeen in the data centers, let's beak quyt, frank right, we have half amillion customers. We really built our name on running data centers reallyefficiently. You know we combining ten twenty thirty servers onto one piece ofhardware. Now there is a lot of enterpise customers that are investedin that platform that do very well on that form, and so, when they take alook at aws and the capability of moving someofe their functionality tothe cloud, that's a long road. The Second Watchteam knows better than me what that road looks like, so it's not easy for enterprisecustomers to move over there, and so, when you have an offering like thmwherewhen your platform on both Onhan in in a WS on the vmdk platform, then it's amatter of moving that workload and actually being in the cloud you can runit. You can move itto. The cloud I mean some people want to call it a fourthlift. Okay, I mean that's, that's probably close to it. It's close, Iwouldn't say it's exact, because there's so much more on thestack than the cloud and there's so much more functionality in the cloud,then customers can build those on their own. So that's that's the thing that I thinkis people don't realize that when they do take on the service they're gettingmore than just a traditional BM mor stack they're, getting that entirestack, all the functionality of NSX and B sand, not just bpar and the latestadditions to at their fingertips, it's Great Francis. So what we see with a lot of our customers?Are It maybe they're new to the cloud, but they have a lot of deep experiencewith vmwhere you? What is your position on, let's say, taking those existingskills that people have on Prim? How does that translate to getting e intoTMC on as yeah? That is one of the advantages of this particular solution,because Tepairon cround is the same as...

...the bstfair in a WS. Now the differencemight be that the version that we run in aws is the latest version of thesspher, but it's backward's compatible. So once the customers are using theservice and they open up that traditional management interface of theB center they're going to realize this is the same thing as what I have onPrim. It looks the same. It feels the same because it is almost the sameother than the drivers for the WS. You know hardware, so they can operate thesame way. They know how to use content libraries. They know how to usetemplates. They know how to measure the CPU. They know what it looks like feelslike that feminate, that familiarity allows them to ramp up and use the service much faster, and I think the one thing that we havenoticed to is once the customers try it out and they realize oh yeah. I canoperate it this the same way. I operate a data center on PREMMAS. We we discover they really run awaywith the service they're like okay, we're ready to go, let's open up asecond location. Let's, let's move the work loads up to the cloud. We've gotan exit off of this particular data center ar off of this Colo right. Ifthey have contracts coming up, that's what We'e see in Francis as well andour experience so far with the Grad segue into an the next question. How doyou think you know it changes that dynamic really from a Dr Perspective,but in a positive way right? What is some of your we'll say, vetenarystatements from a VR perspective and then just your experience working withcustomers here with this in the club. so Dr Becomes you end up with moreoptions. I will say that, but those options right now today, they're not asthere's not as many as they are for built for on premise today. Right so Imean we've got how many years just decades, ind decades of venditors andpartners that have provided Dr Services in theshape of maybe Colos, maybe arental larat. Maybe software, maybe back up software. That can also do Dr.the onpremise ecosystem is very rich, and we understand that. So when you goto the cloud, you've got to work within the constraint of what's available inthe cloud and there are it's a different construct in the cloud. Soone of the things that oure customers are learning, although they know how touse TSE fear, is Ey, it's a different coss model on the cloud. What what doesthat mean? There are different line items, so you want. You want to do, Drto a different location. It's a different cost model than just buying apiece of Software Eh. So there there are some things that there's somebumpers so to speak. That are out there in the Clou today, and I expect thatecosystem to actually grow so you're, going to get more options and it'llbecome more clear, hey skip in Frances. You guys make a great plant with Dr,but in light of where we are in currentafars Covid, nineteen, maybeeighty to ninety percent of the word forus all of a sudden withit a week.You know starting to work remote things in companies and businesses andindustries that have never been remote talked about education, talk about call,centers talk about banking and fihnance. I mean we have literally organizationsthat did not plan both from a capacity standpoint within their data centers,nor from an availability, standpoint and scalability standpoint, and youknow a couple of the use cases that were brought up Duri. The researchwhere DESKTOPS DESKSTOPS TAT may have been supported were desked up as aservice that they hade been supported out of their data center as well. Asyou know, consumer facing properties...

...that weren't set up to scale today arethose areas that VMC, as can support yeah. That is a big use case for ustoday. We are seeing incredible interests in the platform to deliverdesk tops because when customers, even they have delivered desktops on premisefrom their current data center, they bought and they size for the workforcethat they intend to provide promote services to that's all changed intoday's landscape. Like you said eight, ninety percent of the employees aregoing home and have to work remote, but you know what their data center andtheir VDI farm wasn't built to scale for everyone. Only for specific folks.So now they've got to go. They can't build another data center right, quick.They can't just BRACKINSTAC servers and think they can order it. You can't getit up fast enough. So now they've got to go look for options where thathardware and that data center is already available and that software isavailable to serve their the remaining workforce. I think the other case thatwe're seeing is entities that have smaller online presence, whether awebsite that is not commonly used or services that you only need once in awhile, now they're being hit really really hard, and they don't have thecapability to scale on premit. To serve those customers, and so now they'relooking for options as well. U, it's great, you know, I think what youtouched on to is just that he whole procurement model for getting morecapacity and that that is one of the I think most amazing things even prior towhat we were dealing with right now and well. We're all recording this from ourbunkers essentially, but you know so, let's talk about you know precarementDrilla, just a little bit more, but also how it changes like o the hardwarerefresh and life cycle, because I think that is a really moving from you knowfrom a financial model of a capx to an OPX, and I think, that's Fastat. Couldyou talk a little bit more about that as well like that from the opax o thecat backs model that customers are facing today, yeah? That is afundamental change that when you take it to it, which is traditionally workof a cap back model now they're having more conversations with their financialstaff buth their CFO and asking whether their business, the way they generaterevenue can accommodate an oddbacks model if they want to get away from thecapbecks model- and this is all very individual to the companies themselves,but they're seeing they're seeing that value, because if you look at somecompanies that don't run their own data centers, maybe they went from somewhereelse. They have really big contracts, they're, not just one year contracts orthree five ten year contracts for a location to how their data center ortheir Dr Location. Those are big, expensive, con contracts. So if they goto an abex model where they use something like vmc, they can spreadthat cost out and they can figure out. Well, you know what they don't evenhave to upgrade the hardware. It's not their problem. That's my favorite line.When I talk about VMC the Hardwareis, not their problem, the mainenance isnot their problem. The upgrades are not their problem. How much does that savethem in terms of head count? Man, power, man, ours- all of those things comeinto play so I's a conversation between it and their financial staff. Now, whenit comes to that model, t a such a great answer to that question, and justin that so e, do we see companies using this as a transition from that from theold Capbek to Opex? And where you know a lot of our customers are trying tomove to your it as a service as well...

...right. Do you see that, as this being areally catalyst to those kind of efents big time big time? So I've been sayingpart of the reason why I joined BM whereis. You know Wynti to help mycustomers, because, quite frankly, there's not enough technical people inthe world to service the needs of the various companies globally. We've gotto come out with some of these services in some of these products that take theman ours off. Our customers play their staff, they're not growing their stuff,and even they are who are they going to hire from there's? Just not enough ofus to go around, and that is a big determineat for a lot of the managersand the directors that are making that decision when they look at the VMCproducts, and they realized talked about that skill set, it's the sameskill set. I don't have to hire somebody news to learn this. I can runa portion of this somewhere else and not have that cabex model, and I cankeep that head count the same as what it is today, and that is a big piece ofit because what they have seen in the past. If they looked at the cloud model,I don't care if you're going to Zou, GCP or aws, they had to hire somebodyelse or ask the current person to increase their skill set like thatcompelling argument right there for sure. Thank you now, I'm kind of atechnical nerd. Sometimes I like to play one on TV. So one of the coolestfeatures I think in my mind, is you know the way that it gets set up in aws?Is it's a MVME backed B sands? And it really does you know credible speed anda lot of dduping a compression. Can you talk about some o? Maybe your personalfavorite features of the technical platform, the first one that impressed me themost- and I know this is a little odd- is when we closed out pocs like if youwere to close out of POC wuld, say: Don't say they sent you a triand buy apiece of hardware when you're. First of all, you got to figure out when you'regoing to end that POC, okay, the POC's, done you've spun everything down. Well,somebody's got Ta UNRAP that thing unplug. That thing find the originalbox it came in and then ship it. When I turn down anSDDC and BMC, I click like three buttons man and I done. I was like wow that impressed the daylights out of me.I was like we're Dong guys, not the billing stopped the pilling STOPPD,because that's a big dealright. If Yo I mean, if your son had O had a cellphone you're like I got to cut off your cell phone you're like how do you turn that off?You got to call them in advance, you're, like okay plan, to turn off on theThirtiet I'll pay until thethirtye etce. Now this is this is flit over done.That was my favorite feature so far, some of the other ones. I like the ability to add a host like like we talked about. You know youinvest in the data center, you buy pieces of hardware, you ammerotize itover ter or five years. So if you need to do some capacity planning and needto add to that, you've got to budget for it, you've got to get a quote, get aestimate. You got to get a Po Yo got I mean on and on and on it's like six monthbefore you get your piece of hardware and then you got a rackand Stackin andhave to be beg the David Center Guys Tho give you aport. So when you add apost in VMC, you click Abon. How many do you want nine minutes later? Youhave another host addit to the cluster. That's it. If you don't need it, you could turn itoff and you'll only be build that month or whatever. How long? However long youused it like that's, that is easy Pz. In my opinion. It'sgreat appreciate that I stuff. What do you see? I guess just from any majorobjections out there, Francis that you...

...see when you start talking about this.Does anyone have like a pause? You know big pause moment and you know how doyou overcome the objective you know I ea we always weigh it out. Is it's agreat way to dip the toe in the water? As you were just mentioning here, youknow for a pocs and what have you but just curious from your standpoint, whatyou've experienced and you can speak to that a little bit. I think the greatestobjection and I'm not going to hide from this because I've I v dealt withTDIS. Ance I've been with VNR, it's going to be cost BM mores a premium brand. We spent alot of money on engineering on building these kinds of solutions, and I don't know if I counter it as much asI would say: You're looking at the cost of when yousay, coses an objection its because you're looking at it at solely from acompute cost now bif bi said all right. Well, if youhad to do the equivalent you're going to pay for real estate, you're going topay to pull eighteent line, You'e Goin to pay for hardware you're, going hapay for sophmare maintenance, your getting paid for the head count thatyou're going to manage that and you going to pay for their travell cost. Soif you take those costs- and you calculate it for your company now doingapples to apples comparison, that's good point in conjunction your lastTaveen to around just you know, shift into that model of operational cost.You know that you have to figure in the whole iceberg right. So that's reallycool yeah! That's a great way to position that o well and the reality isin today's, like it's also the scale down right. So, let's say if you had amajor Arlaner or even you know, oil gas company that needed to scale downquickly. It's a lot easier to do it if you're on the cloud or on DMC, then, ifyou're running on Pram, going back to your pretious example ofthe POC as a great example yeah, and I think one of the things that sometimesas technical people we lose sight of- is that a lot of our customers they'renot in the technology business. You mentioned oil and gas, but hospitalsand government. They are not in the business of being. You know, they're, not in a technicalbusiness. They just need Tho stuff to run. They shouldn't have to figure outso much of the plumbing and that's why, when I tell customers, I'm like Lhok,the Harwor is not your problem. The plumming is not your problem. The onlyplumat you got to do is from your own premise and how to get in there so tospeak, but upgrades ar not your problem, like none of that is your problem andthey're like that's what they hear because they're like I got to focus onbuilding a hospital, you know doing the oil. I mean they god to focus on thatstuff and they should be. They should not be spending so much mon money andtime and effort into reinventing the wheel. Technically, it's the billversus paying model as be like to say that's great, an for a second. If youcould talk about tools that are out there for analysis in order todetermine how to it go ind what the cost would be. I think a lot of peopleare, you know the thinking about it. They're, like I just don't understand.You know how much hardware I'm going to need how much it's going to cost tooperate. Can you talk a little bit about the ways that Eceta customers forthis could actually get that data ahead of time? So there's no surprises. YeahYeah, you know vias surprisingly transparentwith this particular service on our website. We post AU rogue map, so thefeatures of what's coming in VMC. We also put a a sizing tool online, soyou could go online on the website and you know you could take a subsegment.You know what you think is representative of your workloaw,because not all the Worklov oftentime would be the first thing to move up tothe cloud...

...and and do that cost calculation orhost sizing calculation, and that calculator will take into account allthe software overhead, as well as the size of the nose up in aws, to givethem a rough idea of what they could fit on the hoses in VMC. The other tools I will point to forsome of our larger customers is if they have some of the VM, where managementtools like operations, manager or networks insight. Those will give themvery fine, prannular capability to estimate what how much of the VMCservice they would need for a subsegment of their worklovhes. Sothose are operations. Manager will do that network insight will ha help them track.The traffic flow, because that one is a big one, that people really for for a lat of a better term. Theyreally take for granted when they're on Prim verses when they're in the CLOB great.Thank you for that about using it. As a have you seen, clients out there usingit as a Dr target for just starters, to get involved with or guess what what your been exposure. Wehave one client that has done it that way and just curious from a biggerscale, yeah what you've seen out there so oftenti yeah. We do see, Dr as atarget for Gianci, especially for larger customers and for customers thatare looking at Vdi, failover and birst capability that they will use. Theywill starting VMC in that manner. So it's not headly utilized when theyfirst start out so to speak, but for the smaller clients Dr in VMC, might ormight not make sense. Just because it' he size versus cost model. But ifthey're doing a business connuity case where they put a pilot like using themcasa pilot like then they canlater on the Dr Solution Very easily and just check off that box, so forsome of our smaller client, if they are able to use it in more thanjust a Dr Fashion, they actually are able to check off a lot of boxes. Theycan thun off the cloud strategy. They can check off the pilot life, sobusiness continuity. They can check off the Dr box and potentially the thebursting capability for wor theyre on Prem. So when they have a couple ofthose used cases, then Dr just makes all the sense in the world just tocheck that box. Yea Love it great. That's alsfem! Thankou! U Very MuchYeah! Well, Francis! We try to keep cloud cunch under thirty minutes, soyou know really enjoyed the conversation today. Thank you forjoining us and sharing your insight and we do enjoy working with you as one ofyour partners on these solutions for our clients. But thanks for JoyhisTodan thanks. This was fine. I hope this is a. This is Gidby guys, yeah Yoidn't skip. You know thanks forbeing here as always and listeners. If you have any feedback, questions orcomments, please so Wut that to cloud crunch at Secowachcom, and we lookforward to Chatway to next week I ke e, just thinkyou jok. Thank you Frances.Thank you. Thanks faving me, you've been listening to cloud crunchwith Jeff Aden, EAN, willoughby and skipberry for more information checkout the blog second watchcom company Blok, or reach out to second watch O'nTwitter.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (30)