Welcome to VMware Cloud on DellEMC! For those of you familiar with it, some of this will be a refresh with some definitely new stuff mentioned! And for those of you who have no idea what’s going on here, welcome to this primer!
What’s VMware Cloud on DellEMC?
That’s a good question, and probably a good place to start!
VMware Cloud on DellEMC is a fully managed rack based on VxRail Infrastructure, where everything comes Pre-Built, Pre-Installed, Pre-Configured leveraging the VMware VCF software stack, delivered “technically” as a monthly “cloud-like” service subscription that is all-inclusive.
I’ll dig into quotations and some details later.
This should give you a bit of the gist of interoperation of the various components. Important distinctions here being, YOU control your VMs (and containers) but the pains of management, patching, updating, lifecycle, troubleshooting, firmware, blah blah blah, all the fun stuff… That’s handled by the VMware Cloud team, so in the event a new CVSS 10 Critical Vulnerability should be disclosed for example, you wouldn’t need to directly deal with it as the VMware team would have your back to fix it, as it’s their responsibility!
What’s nice is how the service and support model is handled compared to how things would be traditionally handled On-Premises. Instead of having or needing a dedicated TAM just to jockey your tickets while you try to troubleshoot your latest NSX-T or very weird and anomalous vNIC issue, the VMware Cloud team will be jockeying those requests up the ladder of VMware Support! Also, should any hardware components fail within the rack, Dell EMC Support will be on-site in 4 hours to replace the failed components! (4HR assumes continental US from my tracking)
Y ON THE PREMUS I WANT CLOUDZ
I know, I know, that’s what I’m thinking! Cloud, cloud, cloud! There’s a million reasons to want cloud, “Increased Agility!”, “Simplified Operations”, “Accelerated Innovation” and so much more. And with the advances in the Clouds and their support of security requirements, regulatory, compliance and so forth (look for my coming “Deep Dive into Cloud Regulation coming soon :)” ) But sometimes, there’s some solid reasons to still maintain a physical On-Premises implementation, Risk management, Cost Control, and Increased and controlled performance to name a few.
These tend to be some of the primary reasons people will often lean towards maintaining or establishing a On-Premises infrastructure, even in this day of Cloudify everything! One special call out of this deployment with VMware Cloud on DellEMC is, since this is on-premises, you’ll have a consistent Monthly Service Charge that is all-inclusive, Your cost last month is the same as your cost next month is the same as your cost in 13 months. Contrast that to a burst storm rolling into a micro-service server-less component that keeps initiating API calls to write table state updates, and suddenly experiencing an “unexpected $1,000,000” charge one month. I mean, that’s never happened, forget about it… 🙂
HOW BIG IS THE CLOUDS
That’s a great question, just how big is this… This is where those who are familiar with it might say, “Yea! I know about that!” and those same folks will also get a little envious at what’s new as well! So let’s get racked!
What’s really cute is, it includes a baby version, the “R1” and it’s bigger brother (sister? Do clouds have genders?) the R2!
Okay, sure they’re just a bunch of pictures, you want the meat (or vegan meat option – I’m vegan I can say that) of what’s IN these Racks!
This is what we’re looking at will be beneath the covers of EACH of these servers contained within the chassis! So if you’re looking at the R1 rack, you’d have 3 (up to 5) of one of these servers you specify, or go the R2 route and obviously go with the M1d.medium, because NVMe! And then pack 16 of those into the rack and get 12TB of RAM with all of the other goodies involved!
Sometimes we have to ask the question of, “Why go small, when I can go biiiiiiig!!” And the real reason is, sometimes we have a practical application for a smaller form factor. Let’s say I had a number of applications that required local proximity and were protected for various reasons I could not use Cloud, whether due to distance (Power station out in the boonies with a screaming fast 56K uplink that barely negotiates at 14.4) a storefront operation that calls for a small footprint or really any number of reasons that I am being very careful to not disclose publicly some VERY specific but protected use-cases that are geographically or application sensitive. Needless to say, whether going small or large, there can be some huge benefits here in this compact delivered solution.
WAIT, ISN’T THIS VCE? OR VBLOCK OR VxRAIL?!
That’s a great question! To answer it very easily, it actually IS VxRail (Notice on the last picture the Chassis Form Factor specifies specific VxRail part numbers) as far as the other pieces of it, it’s much closer to the Converged Infrastructure / Hyper-Converged Infrastructure story than it isn’t. This is the direction and narrative vendors have been taking, providing that ease of simplicity and delivery model. But I will say, some of the differences of this solution over the VCE or VBlock model of yesteryear (or even of literally today) is the ease of ordering and deployment.
In this “Sample” ordering scenario, from Order Creation to Live took just under 2 weeks! Yes you’ll say, “I’ll believe it when I see it” and I’ll be with you on that, but I remember build discussions for a Vblock would often take weeks/months, with a very long time to get the delivery even scheduled, so that’s definitely +1 to VMware in this camp!
HOW CAN I GET ONE?
That’s a pretty practical thing to do, right now I want to take you through some of the pieces you encounter during the actual “Ordering Process” some of which are pretty ‘normal’, and some of which I personally feel are PRETTY COOL and things we either take for granted, or often forget about them until it’s far too late and then we’re left dealing with it! (And genuinely, I’m taking you through this mainly so I can show you some things I find really cool, otherwise why even bother showing this process, right?!)
This looks pretty self explanatory, rack size, number of hosts supported, type of networking… TYPE OF POWER?! Yea, that excites me. Anyone who has ever had to call the Power Guy and be ready for bringing up a new circuit or a rack in a datacenter has often dealt with the question of, “What Power Connector Type is it? Single-phase or Three-phase? 110V or 220V?” This is something some of us take for granted, or just hope for the best, but then if you’ve ever been sitting there beneath a rack staring at the plug while yelling to someone, “What kind of whip is it!” you’ll really appreciate it…
This is quite possibly one of the simplest things to do which have almost never been done most of the time. What can I say, it’s the little things. 🙂
It gets even easier. You pick the TYPE you’re looking for, small or medium, number of hosts and you start to make the magic happen!
In case you’re wondering, “Wait, what do all those numbers mean?” That’s a great question, because I often wonder that myself.
Here’s your fun “Secret Decoder Ring” which I’m pretty sure won’t tell you to ‘Drink more Ovaltine’, but should give you some ease of ‘wait, what did I buy?!’
Okay, Okay, this has been easy so far! Will it get hard? Well, kind of. If you have a good relationship with your power guy, get ready to leverage your Networklationship!
This seems pretty straight forward, and that’s because it is! But… you’re still going to need some subnet blocks, ports open on the firewall (TCP:443 and UDP 2426 to be specific for the VeloCloud Phone-home for VMware to manage things!) But you’ll get some great perks with this minimal effort involved in networking. Such as NSX-T configured and ready to go OUT OF BOX! And Micro-segmentation READY TO GO OUT OF BOX! It’s all very exciting as on vExpert NSX to another, after all!
And after that, all that’s left is to choose… your contract term!
Remember when I said this we would discuss this later? Yea, well, it is later! As you can see things are broken up in 3Yr, 1Yr or even a 2-month pilot option (which we won’t even bother to unravel) As I personally care about cost dynamics and ROI and TCO, I’ll have to dig deeper to get the answers of exactly how these pieces of the puzzle will break up, but just like any of our existing Cloud services where we say we’re gonna go monthy (sort of) but instead get a 3Yr contract up front in order to control costs better, it’s more similar to that than it is different.
But at that point, we pull the trigger, check is in the mail… and order away! Seems pretty simple, right? Frankly, it’s probably the easiest of the options we’ve had at this point, passing through multiple iterations of the original VCE family, and VMware Approved Designs, and the EVO-Rail creations, and VSPEX BLUE (Anyone remember that? :)) and VSAN Approved architecture! But at this point, we may have finally reached something that’s a tad bit easier to order, deliver and deploy!
Okay, one EXTRA bonus from the Configurator I’ll share is.. the process to order additional hosts.
This is actually as easy as “Select the cluster I have, which locks me into like-servers”, then “Select the quantity of servers limited to the size of rack” And then, the infamous, “next-next-next-finish” basically. It’s that easy!
Wow this was a lot, anything else?
Yes actually, there are a few things I’ll mention, some I’ll just recap on, but some I didn’t mention at all and I find them VERY exciting!
- New R2 with Single, Three phase, floor and ceiling whip exit! (Up to 12 and 16 host options)
- New Host Configuration: M1d.medium (Dual Socket, NVMe Storage!)
- Expandable capacity – Self-service (ala last picture!)
- LOG Intelligence (LINT) integration!
- Data Recovery Options with: Dell EMC PowerProtect Data Protection Certified AND Veeam Backup and Replication CERTIFIED!
- HCX for Bulk Migration (In tech preview)
- Deployed OUT OF BOX with NSX-T and Micro-segmentation ready!
- Each rack includes a “Standby host” (not counted in your server count or cost) used by VMware for support, failure, failover, and more!
- Predictive Bills without Utility surge charges (unlike VMware Cloud on AWS or other Cloud solutions)
There’s so much more, but this should give you a lot to digest at the moment. This is a great option and a great solution for those of us who have various regulatory, compliance, geographic, data sovereignty, or other reasons, while also giving us an OPEX model versus the traditional CAPEX model of hardware acquisition, while dealing with software depreciation in our operations schedule. Needless to say, this is something which gives you options and is worth taking the time to unwrap the financials of the solution compared to your traditional ‘buy it, build it’ kind of solution we are so commonly used to. I know this is a conversation and series of discussions I’ve had at length for many years.
What do you think? Is this something that you like and want to know more about how it works beneath the covers? Or do you have concerns, reservations of how things operate and how to incorporate this into your DevOps model? Or whatever other questions may roam when trying to ingest these latest updates to the “VMware Cloud on DellEMC” solution! Leave a comment below and I can knock some heads over in VMware Product Management to find out! Thanks and stay safe!