GitLab Product Marketing, 1 Jul 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: 2020.07.01 - GitOps - SDR Enablement Series

Description

In this video, William Chia and Eirini Pan speak about GitOps and our GitOps marketing campaign.

A

Oh all, yours guys, hey everyone, so we have today two very special speakers and William. Are you prepared to present? If not, then we'll we'd like to start off with you yeah.

B

That sounds great I am excited to be on with everyone. My name is William, I am part of the product.

B

Marketing team and in particular, my areas of expertise are cloud native application, development and get ops, and so today we are talking about get ops, which is a bit of a newer term, one that you may have not heard before, but we're kicking off a spectacular get ops campaign led by Rini, which you'll hear about just a little bit so I would love to just give you a little bit of the background of hey what is get ups and kind of who are the personas involved, and maybe some some language, so a basic understanding of what it is and some other resources.

B

You can go to dig in a little bit deeper and then we'll talk about the campaign, so I will share screen here and the main resource I want to point out which should be in this issue, and it's it's right here. What is get ups in these case overview- and this is our get ups- go to market use case page so as kind of a refresher we have about seven or eight use cases. These are like ways you know when folks are looking to buy, get left to do something.

B

It's some kind of use case like dev, suck ups or CI or version control, collaboration and kynapse is something that one of our buyers that's buying. Gitlab would use get lab to do so. In a nutshell, you can think of get-ups as taking DevOps best practices used for application development. Things like birth control, collaboration, compliance and see ICD and applying those to infrastructure automation. So if you think of you probably have seen this slide a lot, we have this at a bunch of different decks and this is the the gitlab flow.

B

So when you're doing application development, you would write your application code. You create an issue. You could emerge request for your code. You commit your changes as you're. Building that application you'd run CI on it, etc, etc and approve those changes.

B

Then you merge those changes back in and now you have that that application that's running in production well in order to run an application in addition to the application code and needs to run on some kind of infrastructure, so it needs to be running on a computer somewhere or a server, or it needs to be running on virtual machines in the cloud or needs to be running on kubernetes.

B

It needs some kind of infrastructure that that application is running on top of, and so the application code is the concern of the software developers in the development team and that infrastructure is the concern of the operations team. This is kind of where you can you've heard of dev and you've heard of ops. We talked a lot about the dev part. This is the ops part. The Ops people are the ones that manage the infrastructure. Now, in the old days of managing infrastructure, this was a manual process.

B

You might have to spin up a physical server literally like rack, a new server on the rack in order to to provide infrastructure for the application scene, but in these days with cloud infrastructure, everything is automated, with these days with kubernetes we're doing dynamic infrastructure that scales up and down. If there's a lot of load on the application, it just scales up and scales down and managing that really really complex infrastructure require some more sophisticated practices in your operations.

B

This is where get ops comes in, get ops is taking these kind of best practices doing things like keeping all your configuration is code. We call that infrastructure as code using version control and and code review and merge requests for your changes. That's kind of what get ops is so so this is. This is a good page.

B

It has a lot of good info I'd, recommend kind of reading it in more detail, but at a high level, if you're doing infrastructure is code or you're, you defining your infrastructure, not manually you're, not clicking in a GUI but you're actually defining that as code, and there are certain tools for this things.

B

Like ansible and terraform and other infrastructures code tools, if you're, storing that in a git repository and then you're using merge requests as your change agent, so rather than making manual changes to your infrastructure rather than logging in to a particular server making those changes, it said you are having C ICD automatically apply those changes when new configuration gets merged into the production branch.

B

So it's just like doing application development, but for your operations, hopefully that's a cleared up of an understanding, and if I can explain that better or if there's questions we can dig into it. There's a few things to key in here. One is just like cloud native. The people you talk to on the phone are probably not familiar with the term get offs.

B

Get-Ups is a great term for us to use in marketing it's a new term, it's a term that we want to gain ownership of, but it's not one, that's very ubiquitous in the in the industry. So if you've heard me talk about cloud native I usually will tell you your customers and the people, your prospects, that you're talking to they're, probably not familiar with the term cloud native, they might be, but they're, probably not they're, probably more familiar with something like kubernetes. So if you hear words like kubernetes and micro services, that's the language.

B

You should use on the phone to talk about cloud native. Similarly, there may or may not know about git ops, probably even less so getup doesn't even newer term, but what they are going to talk about is infrastructure as code. So if you have prospects and they are in the operations team, they're doing platform operations, they're like director of IT, well look at the personas here and just say, but the term that they're that that you would talk to them about is Oh infrastructures code. Are you man? Are you using infrastructure as code?

B

If you are, we have a lot of capabilities that can really help you out in the same way that git lab is amazing for development teams. Your lab is really amazing for operations teams as well and can do infrastructure as code and get lab infrastructures code is really it's a subcomponent of get ops if you're doing get ops, you're doing a bunch of other things in addition to infrastructures code, but that's, probably your term that your prospects are gonna, be most familiar with wanted to give you some language to share there.

B

I won't go deep into all this stuff. I realize it's just a short overview, but I do want to talk about some of the personas here, and this is a slide here and on the stack is also kind of should be linked to the materials.

B

This is a very modern way. Destruction and operations team, so in the very, very old way of you have dev, and you have ops and they're very, very siloed, and you rack and stack servers.

B

You know that's a very old way and there's a lot of different ways to structure operations teams, but this is a very modern way. What we tend to see is there's some kind of what we call the platform operations team and they manage a horizontal platform, usually built on top of kubernetes. That's why the kubernetes logo is there and it's powered by some kind of infrastructure.

B

You might have bare-metal servers that on-premises, like Dell, does that you might be using a cloud platform like Azure or Google or AWS, but the infrastructure is abstracted away from your development teams, your development teams. They don't have to worry about how many servers their provisioning to run their application. All they do is they package their application or their micro service in a container and that container gets deployed onto the kubernetes platform.

B

That platform is managed by an Operations team powered by some infrastructure, and these may be the same team or they may be different teams. Sometimes you just have an ops team and they manage the platform and infrastructure. Sometimes you might be talking to somebody who's, let's say a VP of infrastructure and they just manage this component. Sometimes they'll manage both components right. So these are the kind of that personas here and then even within the dev teams.

B

Dev teams care about get-ups because they care about having a you know: infrastructure provision for them, so you have developers and what we call the embedded site, reliability, engineer, they're, working together in a cross-functional, DevOps micro service team, and they you know this is where they're writing their application code, and this is the infrastructure that their code gets deployed on the teams that manage it. So some of the user personas here are going to be like a site, reliability, engineer, a DevOps engineer, platform operator or operate platform operations engineer.

B

This is the really the big one for get ops if they, if they're doing platform operations or if they want to do platform operations. That's what get ops is all about. Infrastructures code managed as a platform or even somebody might have an Operations title like a system administrator if they're a larger organization and they're, having really embraced engineering further operations, yet where's. The buyer personas here folks, like VP of IT or CIO right infrastructure.

B

Anybody who has infrastructure in their title or again, the newer term is platform platform engineering. So those are some of the personas some of what is get ops and I.

B

Think I'm, just I think I'm, just gonna pause it there. This this main page is that is the one that has like the most info for you. I'll share one last thing and that I really want to turn over to Rini, which is a lot of these go to market motions that were defining things like dev, second ops or CI, or version control and collaboration.

B

Well, we've done is we've developed the entire go to market strategy first and then we start executing against it with get-ups we're doing a little bit differently, actually we're still in the process of defining the go to market strategy, but we're already kicked off campaign so in some ways we're kind of doing this in parallel in some places we're playing catch-up in some cases.

B

What you'll see is there are, this page will be rapidly changing and more content will be here so you'll see, for example, things like our differentiators aren't even defined yet so we've already kicked off the campaign, but we don't have all of these things documented so, which is okay, because we said get ops is so great. We want to go ahead and kick off a campaign right away. We don't want to block and wait on having all the pieces put together so in true gitlab fashion.

B

We're doing this in an inert, MBC, minimum, viable change style, we're we're just starting out and you're, going to see more iterations to come, I headed over to Rini and then I think we'll do some questions at the end. Is that sound good.

C

Thanks William hi everyone, my name is Irene e I am the marketing programs. Team I am joining you from Athens in Boise today and, as William said, get up to the very new use case. We went straight on with campaign. We didn't have much content, we're creating this on the go, but the campaign is live, so I will walk you through the elements just going to share my screen.

C

C

Use my screen perfect. So this is your single source of truth. When it comes to resources for the get-ups use case campaign, there is a link to the parent, epic, our all the work is being done, but most useful links for you are the ones that are listed here. So first of all, I wanted to share something I'm sure fancier and to show you like the campaign journey.

C

So and if you click here where it says campaign journey, you can see this big year. Maybe this is not as big, so this is how people will land on our campaign landing page. So we have some digital ads, live and social ads live and they look like this and they they land on this landing page, which is this one. So our main offer for this campaign is live panel discussion, that's happening on July 15 and we're talking about the future of infrastructure, automation and gear ops. How to do get out.

C

What successful get-ups looks like and it's more of an awareness base, and we have some great panelists from partners and and guests here, as you can see that we'll be speaking on the topic, that's happening on July 15 and that's our main campaign landing page. So our ads are driving mainly to this landing page, so everyone who registers will get +15 lead scoring and then, if they attend live, will get +15 as well.

C

On top of that, if they watch on demand after July 15, they will get plus 15 as well, and then we have more assets, I'll go through them in a bit, but everyone who is consuming Europe's content. They will be entering a energy stream, so we have Pat factory and we have built a path factory experience which is a place where people can consume content. That is relevant to a topic.

C

So we have built this per factory track that is around all the get-ups assets, so people can go in and consume content one after the other. It's also linked here so per factor experience, and it looks like this and it starts with like a few blog posts, and then it goes to like some webcasts that are on the topic and the user can pick which one to consume. So this is also how the nurture stream works. So people would enter the nurture stream.

C

They will get a number of seven emails, moving from awareness to consideration and purchase, so the emails are listed here.

C

If you want to have a look, and also they are on the nurture issue, which is also linked through these epic and its email, is linking to a path factory asset so like it starts with some blogs and then some of them are not have not happened yet so there will be added later and it drives to some gated assets, which we only have some webcasts and mainly with both partners at the moment, and we will have an ebook end of July, which will be added here as well and will be another gated asset and they end up on the free trial page.

C

So when they consume these assets on a factory more than 30 seconds more than 20 seconds at them. Sorry, they get another 30 points, only it's going and that's for the gated assets, not for the blogs and then, if they. Finally, if they click for to start a free trial, they'll get another hundreds course. A hundred points are so that's the overview and then so how? How will you see this lid coming through?

C

So you will see an interesting moment that will say on your salesforce view that will say that they watched live or after July 15. They attended on-demand. The panel discussion get-ups the future of infrastructure, automation in a perfect experience and a perfect experience. So that's what you will see and if they reach enough points through this whole journey, they will reach the 90 point, threat, threshold and and become angels.

C

Also, this is thank you dr. Hughes case campaign, rarely register or attended on demand. The panel discussion heat ups, the future of infrastructure automation we are creating. There is a team that is creating now some outreach sequences, both outbound and inbound they're, not ready yet they're working progress, but they will be ready for July 15th. So once so before the panel discussion happens and before those leads get added, you will have the output sequences to follow up and also to drive them to the on demand. Page.

C

Let's see what links I missed here, oh okay, so this is a link to the campaign brief. So this is talking about what the objectives are for the campaign and you can wade through that. I just wanted to show you that there is a list here about all the marketing assets that are relevant to this use case campaign. So we have a topic page, which is more like a wareness about what is get-ups and then it's listing resources, and then we have some blogs. You can see what what's like what not yet.

C

We have a lot of partner webcasts, as mentioned on this topic that are also gated content, and then we have some more webcast embeds duration stage and we're building out a solution spades as well, and we will have an ebook end of July, which is also going to be gated content on this topic.

C

So we talked about persona, William talked about the use case, go to market topic page, we saw a factor experience or so, and there is so they did. This is ongoing. As I said, there is another webcast, that's happening on the topic as well on that July 8th and this will be another resource as well. I.

C

Think covered everything see the agenda.

C

Yeah I think we're good to open for questions all.

D

Right I have questions so within civil and terraform. Obviously, they're commonly used in the infrastructure is e part side of things. My question is: what would be the advantage of combining that would get lab comparing to our competitors and we'll get lucky eventually replace those tools? Are there? Are there any interests on the roadmap to to also to cover that, so there is no need to integrate insulin to reform in the future.

B

That is a really good question. I will show this slide here as well, which I linked in the comments. This is not quite a customer deck that asset will be forthcoming. This is more of like a speaker deck if we were to do a talk of what is get ops.

B

So if you look at some of the technologies involved, you need things like a get management tool. Ci, CD, container registry gitlab does all these things, but you also are gonna want things like a configuration manager or infrastructure provisioning, and the answer here is somewhat technical and somewhat complex. So I will say this: if you are container izing your workloads, that means you're not dealing with virtual machines at all, then.

B

In many ways your docker file becomes the description of the infrastructure, and so, if you're, using a docker file to describe everything, you don't necessarily need a configuration manager in some instances, but without getting too technical about it. There are still a lot of reasons, even if you're, using Cooper Neddie, that you do want some kind of configuration manager to configure servers, and you do want some type of infrastructure.

B

Provisioning, which is say, okay, actually go and get me, though, these virtual machines from this cloud service, which is what something like terraform or plumie or cloud formation view so get lab, does not have on their roadmap any type of capabilities like traditional configuration management tools like ansible or chef or puppet. We have no plans to replace the in terms of a traditional virtual machine workflow, and we have no plans to replace provisioning tools like terraformer plumie. What we do have is tight integrations with those tools.

B

So one of the reasons why I get lab would be better than say using a bitbucket or github for your infrastructure as code is that we have tight integrations into terraform. For example, I don't want to get too technical, but one example of a feature is terraformed State. The way terraform works is it's called eventual consistency. You define the state of what you want your infrastructure to be, and then your automation tools.

B

So, for example, let's say: get labs, see ICD work to configure your infrastructure provision, your infrastructure, so that it matches that state and that state is stored in a terraform state file and that's something that, when you're, using terraform on any platform with any technology that state file needs to be stored somewhere well, gitlab will store that file. For you. This is an example of our tight integration with terraform and it even shows up in the merge request.

B

So when you see a merge request that you're working with terraform, you have the terraform integration enabled you can actually get to that. State file is stored on get lab or, and it's stored within get lab, and you lab understands what it is. So the summary would be get lab has tied integrations with tools like terraform that make it better than other tools.

B

That's one summary: we don't have any plans to build our own infrastructure provisioning, our own configuration management, no roadmap, for that. So we'll continue to integrate with tools like ansible and terraform like, for example, we've. No, we have no desire to build our own kubernetes, we'll continue to integrate with kubernetes and then the other advantages here are the same reasons that make like our version control collaboration better or that make our CI better right in order to do get ops, you're going to be using CI CD.

B

So all of the things that make our CI CD better for application development make our CIC be better for get ops all the things that make our version control collaboration, our SCM better for application development are also make it better for get-ups. That's a really good question. Is that does that kind of help you out yeah.

D

So I, wouldn't ask kind of touch from this state file. Is that the feature that we offer the relay studio and disaster in the ultimate tier and the second question would be? Is there any advantage of doing IAC on Prem like a get up.

B

Some Prem, so you can do get ups with self management lab or you can do it with get lab. Comm doesn't matter which one you do. The advantages to going on crown versus in the cloud are all the same ones that you would have otherwise, like things like. If you want to have compliance, you know challenges that you want to manage your own server.

B

That's why you would do it in terms of your get ops, workflow it's going to work the same, whether you're using self-managed or get lab comm in terms of the terraform state administration. Here is a Doc's link. I'll put that into the chat for today and I'll. Try to also comment it on the issue. It is unrelated to geo, or anything like that. So I believe that this is a feature that's available in get lab cord.

B

So this is a value of using git lab generally, so, in terms of you know, get lab premium or get lab ultimate. All of the same other values of those tiers are mostly going to apply with this use case as well. Awesome.

D

Thank you so much yeah.

A

All right guys, not guys, sorry, everyone are there any other questions or, if not, we'll, give you back one minute of your time. Good.

D

Catch crass, look at you.