Episode 22 - vCF Product & Packaging - Part 2
Hello, and welcome to our next episode of the VCF Roundtable. Today, it is episode number 22. We're talking about the VCF product packaging.
Matthias Eisner:You you're starting with a joke, Toby. It's called VCD roundtable. Yeah. VCR. Yeah.
Matthias Eisner:It is
Tobias Paschek:the VCD roundtable. Yeah. That's absolutely right. It is the VCD roundtable. And today, we are in an absolute elite round, because with me is
Matthias Eisner:Yeah. I know. This is Matthias. VCT roundtable talking about VCF product and packaging. Yeah.
Matthias Eisner:We have the VCF roundtable talking about VCT product and pack whatever.
Tobias Paschek:What whatever. Okay. So what is in our last episode, we have already started to talk about the product packaging, what is included in VCF, from a service provider perspective. And today, we would like to talk about the additional products or the additional topics which are there. So let's have a look what we have in the product packaging.
Matthias Eisner:Exactly. So last episode, we talked about SDDC measure, vSphere, v SEND, and NSX. And we said next episode is going to cover the ARIA suite, which is part of VCF, and be more precisely, it's the ARIA suite enterprise edition. So we have ARIA operations for network, and then we have the audio suite with operations manager, operations for automation orchestrator, and, of course, the life cycle manager. So, we're we're not talking about ARIA for networks because that's that's a different ballgame, because what the product is able to deliver and what it's meant to be used for.
Matthias Eisner:So we can well, we could make a a dedicated episode out of it. But first and foremost, what what is about the Alia suite? So it is part of the VCF product package even for service providers. So we could start complaining about, yeah, what are we going to do with it? I don't need it.
Matthias Eisner:I don't want to pay for it. Whatever. It doesn't make any sense because it's part of VCF, and I think it adds a ton of value, which is need to figure out how can we gain the value out of the additional product suite and how can we, as service providers, provide additional services for our tenants. So for example, let's start with IA operations for log. So first and foremost, it's still important to say it still has no built in multitenancy.
Matthias Eisner:We should be very, very, clear with that one. So I would start with with the service provider use case. In the end, as a service provider, I can use it and collect all the log files from my infrastructure. And we put ClioRe's on top of it just enabling me to provide a better maintenance and operations and services for my infrastructure, to support or to proactively support my tenants, if something happens within the infrastructure. But, on the other hand, Toby, so you are you're one of our VCF guys.
Matthias Eisner:So what about a dedicated work automate for instance?
Tobias Paschek:Yeah. I I was just about to mention this because, what we still have also from a service provider perspective is to ramp up a dedicated isolator. And now we really need, as as you mentioned before, we need to be precisely because there are 2 different options available related to an, let's call it still dedicated workload. There workload domain. There is the dedicated workload domain within the my, VCF stack.
Tobias Paschek:So which in the which within the same single sign on domain, and there is the isolated virtual domain. And only from my understanding as as today, I'm at the moment, I'm only capable of having a dedicated, other operations stack or a a other stack in an isolated environment, In a shared environment or in an in in an, normal vcf, VI domain environment, I only can use a single instance. But let's see what Broadcom is doing here and how the stuff is changing in the future.
Matthias Eisner:So so you're saying the difference between an isolated and just the workload domain in VCF is, kind of enhanced link mode or isolated SSL domain of the VCF document. Yeah. Just to get me in sync.
Tobias Paschek:Yeah. Yeah. Absolutely. Absolutely. So, that's one.
Tobias Paschek:But still, it is one option. And to be honest, we have, discussed this already in the in the last couple of meetings and the last couple of weeks. Still, yes, there are some automated deployments in my whole, VCF stack, but no one says you are still not allowed to deploy it manually. So, So, also, this is still an optional, still an additional path to say, okay. I cannot utilize the whole automated deployment as today, but still I can deploy an additional, let's stay on ARIA operations for log, instance for a dedicated customer.
Tobias Paschek:So the this is possible, and this is a working example or is a working would be a working solution. But let's see how the road map is, developing there and what we have in the next couple of weeks next cup couple of months.
Matthias Eisner:Mhmm. Yeah. So the the dedicated instance for for a tenant, I think we need to be careful because I I foresee the use cases, which, can't it can't be used for. It it can't be used for the underlying infrastructure because Yeah. VCenter has one Syslog collector, same for a new 6i host and and same for NSX.
Matthias Eisner:Let's think about, for example, a DFW or or DH transport nodes.
Tobias Paschek:Yeah. Absolutely.
Matthias Eisner:But if a if a customer would like or would prefer having a manage kind of syslog collector instance to reconfigure his guest OSs, for example, to collect the log files. Of course, I know there there is a ton of stuff to configure in ARIA operations for logs because of agents and and whatever. But let's move that. The whole configuration side is just about the use case.
Tobias Paschek:Use case.
Matthias Eisner:So the customer is able to log his OS logs and maybe application logs into the the operations for log instance. The only question we always need to be aware of is, how about licensing. Yeah. Absolutely. Can I ask you today, I think, because, the vcf part contains the whole infrastructure layer, in terms of vCenter and and and all the the VMware by Broadcom product stack?
Matthias Eisner:I have no idea. As far as I'm aware of, as long as the guest OS runs on the same virtual infrastructure, I think it's okay.
Tobias Paschek:I've to be to be honest, also from from my side, not really a clear answer, but what I, have read in the whole new licenses, documents and papers, what are already, available. Not not that much content, but, the first documents are already available, is yes. You are absolutely right. As long as you have your OS instance, your, guest application running on the VCF license or on the VCF license stack.
Matthias Eisner:Yep. There And the virtual machine needs to run on those ESXi hosts.
Tobias Paschek:On those ESXi hosts, you don't need to have additional licenses, which is, in general, to be honest. So as far as we have, understood right now, a total different approach than VMware was, was going in the park. But let's see what, in the long run, really, the solution would be.
Matthias Eisner:Yep. But, again, again, the licensing assumptions we're coming up with is just as we understand. So that does not mean that it is the licensing is done like this. It's just how we understand.
Tobias Paschek:W w w.com.com. Yeah.
Matthias Eisner:What well, what's important if you have maybe a dedicated and isolated work of domain with with a service provider and have your own infrastructure on prem and you wanna have log or you have you wanna have sent all the logs for on premises and cloud to the same centralized ops for log dedicated instance. So that that would be an example. I have no idea how this works from a licensing perspective because the the virtual machines running in the cloud on the PCF license stack, the virtual machine went. So that's that's more complex.
Tobias Paschek:Yep. Absolutely.
Matthias Eisner:And another one, but we have provided a dedicated video around the whole NSX and and projects stuff. So projects within NSX. So other operations for lock together with NSX projects enable small dependent logging view for networking services. Networking again. And firewalling.
Matthias Eisner:And firewalling. Yeah. Yeah. But, again, we have a video created dedicated on this topic. So Anzac's project, advantages, disadvantages about, that configuration.
Matthias Eisner:That's ARIA operations for log. Which one is next?
Tobias Paschek:Our world famous ARIA operations manager. Also, for that, to be honest, we had one session with our Belgium phenomenal, Alan Greenet, about operations manager. So if you are really interested in the operations manager stuff, have a look in the past episodes. But today, we are talking about what is there from, VCD and VCF perspective.
Matthias Eisner:Yep. It's it's it's again a use case discussion. Right? We're aware The ops manager is an enterprise monitoring tool, but what can we do with it? So I think we have 2 different kinds of use cases.
Matthias Eisner:We have the service provider use case and the tenant use case. For the service provider, it's still the same. So we have it because of the core based licensing. The service provider was allowed to use it to monitor his or to use it as a monitoring tool of the infrastructure for his own purposes for capacity planning and and, SLA monitoring and and all that kind of stuff. But he was not allowed to provide access to the tenants from a licensing perspective, only licensing.
Matthias Eisner:Yeah. Yep. So this the service provider use case is basically speaking the same as before. And, again, it gets more interesting nowadays because of that's the first bullet, because of of the core based licensing. In the past, we had the VRAM based licensing, so we had enough host powered on install retro on, and the license fee was paid based on consume vRAM.
Matthias Eisner:Nowadays, we are paying, based on up and running course. So it might make sense to use ops manager to optimize the infrastructure to have less servers running, so to save some money on licensing. It's just an idea. I know we have the difference between committed course and and on demand and then all that kind of stuff, but it gets more interesting to optimize the infrastructure usage to save a bit on the licensing cost.
Tobias Paschek:Right sizing. Right sizing. Yeah. What what we mentioned already in in in one of the past episodes, with with the new licensing model, rightsizing becomes more and more interesting. There is the there I can clearly say that the operations manager will heavily support you as a service provider and also maybe your, tenants to right size your infrastructure.
Matthias Eisner:Yep. Yep. To to increase demand. That's what tenants should do. Right?
Matthias Eisner:To increase the answer.
Tobias Paschek:Yeah. Absolutely.
Matthias Eisner:Consumption. But, yeah, it's it's the second use case because in the past, I I know only a very few tenants willing to pay the extra amount of money to have access to operations manager. Because most of the tenants didn't get the advantages of this enterprise monitoring tool. But with the new VCF stack, licensing is done. Right?
Matthias Eisner:So there is no additional cost to have operations manager accessible by the tenants. So from my point of view as a service provider, you can just develop use cases for the tenants, create ready to run dashboards, which can be used by by your your tenants, enabling them. As Toby mentioned, maybe some rightsizing or, increased demand dashboards, whatever, and provide access to a tool which you're already paying for. And and from our perspective, just being honest, it's it's not if service providers are going to use it and and and add value for their customers. It's just a a matter of time, because as Eve already mentioned in in I think in more or less every single video, we now have a plain battlefield.
Matthias Eisner:Every service provider licenses the VCF stack. Every service provider has access to the same products. So the question is, how can we get the most out of it? How can we provide the best value for for our tenants?
Tobias Paschek:Yep. As as you just meant because I was just about to mention it, the the most important question is how quickly is a service provide from my perspective, to be honest, is how quickly, is a service provider now capable to provide the new additional services because they are there? And as always, the last point gets bite by the by the dogs.
Matthias Eisner:Yeah. True. True. And and and these 2 two products like I operations manager and operations for logs, They being really honest, they provide basic infrastructure monitoring and and and data, which I should be aware of.
Tobias Paschek:Mhmm.
Matthias Eisner:Also What's next on the list?
Tobias Paschek:Also helping you, figuring out root cause stuff.
Matthias Eisner:Oh, yeah.
Tobias Paschek:So the next one is our ARIA Orchestrator. Matias, would you like to talk about it?
Matthias Eisner:My most beloved product. No. So audio orchestrator, it it's a workflow engine. Right? It's there since decades.
Matthias Eisner:But but first and foremost, I would still like to mention, same file orchestrator, it's not a multitenant product. So it's not meant that a service provider gives access to to a tenant directly to to the orchestrator. Even even installing a dedicated orchestrator instance for a customer might cause issues. It it really depends. I think if you have an isolated workload domain, you could do it somehow, but I would not prefer that option.
Matthias Eisner:What what's it doing? It it connects to external system. It's just a huge API. It runs workflows, collects data, updates external systems, whatever. I I still see the orchestrator as a centralized tool to enable even more self-service services for tenants to configure or access as service provider hosted systems.
Matthias Eisner:I think a very simple example if, a service provider hosts a Microsoft active directory providing usernames and log ons for for, attendants, it could provide a self-service like, oh, a reset password because you're not enabling access to the to the central active directory for anyone else. The only very important thing with orchestrator is, please, if you build services around it, never ever provide direct access because Orchestrator has administrative access into so many systems. It's very dangerous. So please expose orchestrator only via the content hub in VCD because VCD solves the whole multitenant issues or challenges we have around orchestrator. That's just please please please do do not provide direct access.
Tobias Paschek:I would say not not only we have within Orchestrate, I would say also we have Insight vCenter and many other products.
Matthias Eisner:Yes, sir. You're perfectly right. But Are the automations?
Tobias Paschek:Are the automations? We we we discussed it already yesterday. Maybe some some SPs will think about the same. Okay. We are now in the phase, that VCF, out of the box, provides already the capability of deploying and consuming ARIA Automation, which is supposed to be a cloud service portal.
Tobias Paschek:But on the other hand, we are talking about, Cloud Director in the vCloudDirect or the CloudDirector roundtable and not in the VCF roundtable. So what should we do with ARIA Automation? Yesterday, we discussed it already. We figured out, yeah, one option could be if still the the same use case we had already in the, ARIA, operations for lock, if you have a customer inside your data center, inside your service provider environment saying that I would like to have a dedicated isolated workload domain, and I don't need CloudDirector for whatever kind of reason. But, maybe he would like to have, still self-service, then he can utilize here aria automation.
Tobias Paschek:Bless you. So this this is one use case. And the second use case we figured out or we provided, it was already there in version 7, and it is now, available again in version 8. Aria Automation has an endpoint for Cloud Director. So, we can utilize ARIA Automation more or less in front of my Cloud Director, to provide services inside the VCT environment.
Tobias Paschek:So also there, if I have a customer or if I have a tenant saying I have a dedicated VI or the dedicated workload domain, but also still consuming some additional services from an shared environment, then I can use here also aria automation to provide, self-service against to all of my endpoints. So this is the use case we we can imagine at the moment. Also here, let's see what the future is really bringing, I would say.
Matthias Eisner:Exactly. So and and that's that's that's the nature of our audio automation to say, I I I build a cloud template, and I just deploy the same cloud template against different infrastructures. And that could be either the vSphere based dedicated isolated workload domain, which I used to host production workloads on. And on the second, I can have the VCD, endpoint from my CSP and deployment test instances against using the same code template, for example. So that that would be a a use case for audio automation.
Matthias Eisner:But, again, very important, audio automation itself or by its nature. And, oh, no. I'm I'm open that kind of hole, that that that wormhole. It it's not fully multitenant. The box
Tobias Paschek:the box of Pandora is now open.
Matthias Eisner:Yeah. No. I I don't want we don't want to start a discussion on that. So it's still, as far as we understand, the preferred option if you have multiple different tenants. Please have a dedicated ARIA Automation instance per tenant.
Matthias Eisner:And that brings the audio suite life cycle manager into the gate because if we start saying half the dedicated audio automation for tenant a and tenant b, and how shall I manage all those different infrastructures?
Tobias Paschek:Yep. And to be honest, it is also part of the of the whole, Cloud Foundation story because the SDDC manager, which is in at the end, the SDDC manager is just triggering the Adios suite, life cycle manager to deploying my Workspace ONE Access, so the identity manager part, deploying my other automation, deploying my ARIA operations, and so on. So at the end, there is already the whole integration between the SDDC manager and the ARIA suite, life cycle manager. And the SDDC manager is just calling the ARIA suite life cycle manager and says, hey. Can you please deploy Workspace ONE, access to the the IDM portion?
Tobias Paschek:Can you please deploy operations manager? Can you please deploy log? Can you please deploy whatever you would like to have? So also here for all products, in the whole area suite, there is already some integration between VCF and the ARIA suite. Not fully, to be honest.
Tobias Paschek:So, some of the, some of the other products are not, there to be automatically deployed from a VCF perspective, then I still need to go to my life cycle manager but can start the manual deployment. As you mentioned already, to have the infrastructure overall utilized and also in a in a, self deployment engine.
Matthias Eisner:Yeah.
Tobias Paschek:Yeah. With that being said.
Matthias Eisner:So so what's the what's the conclusion out of it before we go before we move on with announcements? The conclusion is, we have 2 options. Option a is we complain about the price tag and that we're not using all those products. It's an option, but not a good one. That's Not a good one.
Matthias Eisner:I think. So option b, come up with proper use cases and build use cases around those products, and there are many. What we prefer or what we suggest is build a ready to run product around that use case. So make it consumable. If you approach a customer saying, Ryan, dear dear, mister customer, you wanna have operations manager and monitor infrastructure because, like, yeah.
Matthias Eisner:So how does it work? How how can it consume? Blah blah blah blah. And you if you confront the customer with our operations manager and and that guy has never seen that product before, he might be overwhelmed by the the ton of the amount of data presented, all that kind of stuff. So have a ready to run product like some dashboards you come up with, and and maybe some additional super metrics, whatever is cool within the product.
Matthias Eisner:And have package with a certain flow and and the data sheet and basic guide and or operations manual to present to the customer and then add some managed service on top then sell it. Mhmm. That makes perfect sense. So that's what we suggest to do. But the same for audio operations for log and and aria automation and that kind of stuff.
Matthias Eisner:So that's what we suggest to do. Yeah. So Toby already showed the the last slide we have for today, which is about announcements.
Tobias Paschek:Duplicated information. Is Cassandra still still an option in VCD, or is it already gone? So the old, matrix login is already, has already moved out
Matthias Eisner:from still there. I think you could still use it. I'm not a 100% sure. Yeah. But Because it won't be the way going forward.
Tobias Paschek:Yeah. But but also for for those service providers which are maybe utilizing it still, you have now a perfect option to get rid of it. Just bring in our operations, manager and the tenant app and provide the same or to be honest, much more capability to your tenant from time to time.
Matthias Eisner:What what nice feature with Cassandra, but we're not starting all over there.
Tobias Paschek:Over. Yeah. A little bit of announcements.
Matthias Eisner:So the next PCD roundtable is planned to be released beginning of March. That's in in a week and a half, with, big news on the cloud provider program, which we hope will be released by then. So if if it won't be released, we don't have any news, but, yeah, we'll see. So Eve and Sasha will be at the MSP CSP conference in Vegas, 11th until 14th March. So they're on-site.
Matthias Eisner:If you wanna meet those guys, just reach out. More than happy to help scheduling a meeting if you need something. And in March, there's also a CloudFest Europe in Host. That's the south and west part of Germany. So close to the Swiss border, that's the next conference conference attendances we have on the schedule.
Matthias Eisner:And with the cup or with the next VC round table beginning of March, we might also release or announce already our planned attendances within April.
Tobias Paschek:Yes. So the plan. And one last announcement because Yves and I finally decided the name. There will be a new podcast livestream stuff available in a couple of weeks. So Yves and I decided to, do the same like we have done for the v c roundtable.
Tobias Paschek:We will do it with VCF. Now we are talking about VCF, but more in a really hands on scenarios. We will start, a Twitch series, really of what is the deployment axle, what is the the, cloud, builder appliance, how to ramp up a VCF stack really from from 0. So this will be also available in a couple of weeks.
Matthias Eisner:That sounds awesome. I'll I'll attend. I have a ton of questions.
Tobias Paschek:It will run, under the world famous name of VCF Musketeers.
Matthias Eisner:Oh my gosh. Nice. I I do something with naked cats.
Tobias Paschek:Yeah. For to be sure.
Matthias Eisner:Alright. Alright. Things thanks, everyone, for attending. No questions as far as I'm aware of? Nope.
Matthias Eisner:Best of luck. Have fun coming up with a ton of products and ideas how to leverage all the products being part of the VCF stack. See you at the next VC roundtable. Have a great day. Thank you very much.
Tobias Paschek:Bye.
Creators and Guests
