Guest: Rob Hirschfeld (LinkedIn, Twitter)
Company: RackN (Twitter)
Show: Let’s Talk
Keyword: Infrastructure as code (IaC) 

Summary: RackN has announced Digital Rebar v4.9 which focuses on the company’s infrastructure as code (IaC) at scale capabilities. Digital Rebar creates reusable, standardized processes for platform and infrastructure teams, enabling both self-management and control at scale. Let’s dive deeper into the new features of the latest release with Rob Hirschfeld, Co-Founder and CEO of RackN.

“We have taken the Digital Rebar workflows and opened up the integration completely. Any type of systems that need to get integrated into Digital Rebar can just drop in and do that integration,” says Hirschfeld.

Highlight of the show:

• What are the new features that users or customers should be most intrigued about in Digital Rebar v4.9?

About Rob Hirschfeld: Rob has been in the cloud and infrastructure space for 20 years and has done everything from start-ups working with early ESX betas to serving four terms on the OpenStack Foundation Board and as an executive at Dell. As leader on the2030.cloud, he believes that the technology of running data centers and applications on cloud is just part of the bigger story. He trained as an Industrial Engineer and carries a passion for applying Lean and Agile processes to software delivery. Rob has received degrees from Duke University and Louisiana State University.

About RackN: RackN was founded with the vision of operators working together to improve IT operations through shared software and community regardless of infrastructure type. The company believes that operators should be able to maintain control, consistency and choice over their infrastructure without requiring prescriptive appliances or managed offerings.

Here is the full, unedited transcript of the show:

• Swapnil Bhartiya: Hi, this is Swapnil Bhartiya and welcome to our newsroom. And today we have with us, once again, Rob Hirschfeld, CEO and co-founder of RackN. And then today we are going to talk about Digital Rebar version 4.9. Rob, first of all, welcome to the show.

Rob Hirschfeld: Thank you. It’s exciting to be here.

• Swapnil Bhartiya: Of course, this release is still interesting. If I ask you, what is the expected release date? When are you planning to drop it?

Rob Hirschfeld: Probably March 1st, in the first week of March is our planned drop. Challenge with having scale customers is they expect every new feature to work at scale. So, got to dot all the i’s and cross all the t’s.

• Swapnil Bhartiya: Right. So let’s talk about what are the new features that users should be expecting in this release?

Rob Hirschfeld: This feature really focuses on our infrastructure at scale, infrastructure as code at scale capabilities. And so there’s some really significant ads that customers have been asking for. One of the top of the list for me is webhook, web integrations, triggers, and timers. So what we’ve been able to do is take the Digital Rebar workflows and our customers use them integrated into other processes, it’s just a natural part of how the systems go. But what we’ve seen is that the interest in get-ops and using version control to really drive automation and configuration needs a way to cleanly say, oh, I changed something in my repos, drive the new states into my infrastructure. And what we’ve been able to do is really open up that integration completely. So any types of systems that need to get integrated into Digital Rebar can just drop in and do that integration. And we’ve had for a long time, the opposite direction where we call back and check in and push things to other infrastructures.

• Swapnil Bhartiya: What other notable features are there of this release?

Rob Hirschfeld: So in this release, we’ve taken the final breaks off of our clusters and resource broker capabilities. And also, there’s a new feature in here called Work Orders. All of those actually work together to have this very clear distinction between building and running things. So RackN has had infrastructure pipelines, and the idea that I could do a end-to-end process for as long as we’ve had Digital Rebar, but in this, the last two releases, we’ve really extended the capabilities of having a system that can do running operations, that Day 2 action. So we take the same automation pieces, but instead of having to hook them together as a workflow, we take that infrastructure as code thinking and turn it into locks of execution, what we call Work Orders.

Rob Hirschfeld: And those Work Orders are incredibly portable, they’re very easy to use, you can put them on timers or schedules. We just did a couple of demos for Cloud Field Day, where we showed doing a global patch distribution using a work order system. So just run the patch, see if it worked and then execute all those pieces in parallel. It’s really groundbreaking automation because it’s not new. It’s just using our current automation in a more ad hoc, day to day way.

• Swapnil Bhartiya: One thing about this really, which I was reading in your newsletter was also about, of course, data operation. And then you also talk about scalability. So I also want to talk about how you connect these two.

Rob Hirschfeld: Scalability is one of the biggest challenges and we hear customers complain about complexity all the time. And the complexity problem is a scale problem, because what we see is that people are able to use Terraform, or Ansible, or scripts, or any of these tools team by team. But when you try to do it across an organization where you actually have the collaboration and coordination to accomplish reusable work, that’s where things start to break down. And that’s exactly where RackN has focused. So the way we approach infrastructure as code, is all about how to take and scale those systems together. And that comes back to modularity and reuse and collaboration. So when you look at Day 2 operations, the thing that gets really powerful here is not that we can run a whole bunch of tasks on a system.

That it is important, that’s Day 2 operations, but it’s actually that we can do the same operation that we would do if we were provisioning it. Or we can encapsulate it into infrastructure as code so that operation is a standardized piece of work, and then everybody can see it and you can reuse it. It takes down to do Day 2 operation if you only do it at the team level, right? The goal here is that your platform team or your SREs can actually start applying the same logic and the same work across your organization. And you can’t do that if you’ve got anything that’s custom.

• Swapnil Bhartiya: Of course, thanks for explaining that. Now let’s talk about what else is there that you find significant or important in this release that not only your users or customer, but you yourself are very excited about?

Rob Hirschfeld: So, I’m really excited with the way we’ve been approaching multi-cloud and hybrid cloud. So we’ve had capabilities to integrate wrap, things like Terraform and Ansible for a long time. But in this release and starting in the last release, we’ve added these concepts called resource brokers and clusters. And what those that are able to do, is they are the interface points for teams that are actually doing the real work. So instead of having to worry about a machine, which is not a construct that most people want to deal with, our users are actually able to start working at a cluster level and they have an API for clusters where they can drive operations for a whole cluster of machines or add systems, right? To add machines to a cluster, remove machines from a cluster, create a workflow that is specific to the cluster, even if the cluster is empty.

And this was the thing that really blew my mind on how we’ve approached this problem from a tech perspective. When people are dealing with clusters, usually clusters don’t exist until there’s machines in it. With Digital Rebar 4.9, the cluster could exist with a life cycle completely outside of whether it has resources or not. So that would allow you to do something like spin up a cluster every morning and assign resources to it, grow it, and then at the end of the day, shrink it. So if you have a development test cycle where you have somebody who’s playing with something, you can actually say, set up my cluster every morning when I come in and get my coffee and tear it down every night when I go home, and just have those automated operations. And that works because you’re dealing with things at the cluster level. It’s a totally different way to think about infrastructure because we’re getting out of this idea that you have to have a machine to have infrastructure.

We actually have infrastructure constructs completely outside of it, outside of the actual machines that have been assigned to you for work. Same thing is true with resource brokers, and we’ve seen people get very excited. A resource broker is an interface backed to a cloud system, but the way we design them, it’s not like there’s a generic Amazon or a Linode resource broker. You can have resource brokers that are team specific or individual specific or function or region specific, even though they’re doing something similar. And that would mean that instead of having to funnel everybody, and this is about scale, instead of having to funnel everybody through the same resource brokers with the same choke points, you could actually say, I have a team that needs to use Amazon or Linode or Google. They have their own token credentials, chargeback, billing, permissions, right?

All of that can be encapsulated in a way that allows that team to have access at a known API point. And then you can have other teams with other credentials and it makes it more manageable. We keep coming back to this Day 2 operations and scale. I want to have the teams that I’m supporting have places where I know they do their work, and if they’re having problems, I want to go be able to check on and see what they’re doing and make sure that everything’s going according to plan. That’s what we’ve been doing with Digital Rebar and by exposing these core API constructs for the first time, this resource brokers and clusters, we’ve really extended the control points that teams have to collaborate together. And at the end of the day, it all comes back to, does your software help you collaborate? And that’s what we work on.

• Swapnil Bhartiya: Rob, thank you so much for visiting and taking time out today to talk about this upcoming release. And as usual, I would love to have you back on the show. Thank you.

Rob Hirschfeld: Thank you. It’s a pleasure.