We wanted to jump into an overview of BindPlane and then follow that up quickly with a couple of demos. First, BindPlane first and then of VMware.
So as I mentioned, you know, BindPlane is monitoring integration as a service. So this is a SaaS so this is a SAS solution designed to support all of your data center technologies through a single solution and then send that data along to any SaaS monitoring platform that you have configured–that you’re using inside of your environment. It takes any of the 150 integrations that we have already developed–any of those are available and can be used within our monitoring platforms.
What we found is that if you have a large environment, so let’s say you have hundreds of technologies inside of your environment, which is very common for a large company, and you have, say, two or more monitoring solutions that you want to use, it can be extremely difficult to manage. And not just the initial installation of all those integrations but also the lifecycle management. How do you maintain these integrations over time across hundreds or thousands of systems? How do you integrate with multiple platforms? And for us, BindPlane is a solution to that.
Let me go switch over to a demo here. Yeah we’ll see that. Alright. So this is BindPlane. I’ll walk through a couple of the components here. This is an overview page. What you’re looking at here is three different data centers. There’s an AWS data center. There is an on-prem data center that we have locally, and then on the other end, there’s an Azure data center. Each one of these has a single collector located within it. It’s a single application, that we install, and then it is configured with a variety of different technologies. In this case, AWS obviously, the majority of those are AWS technologies. For our on-premise solution, I see that there are database technologies, there’s hardware. Then for Azure, it’s primarily Azure technologies, also some databases, Kubernetes as well. What you’re seeing is all the data is being collected and sent through the BindPlane service, and then it’s passed along to what we call destinations.
So there are multiple monitoring platforms, one Microsoft OMS. It is a monitoring solution inside of Microsoft Azure. One New Relic, I know many folks are familiar with it. Let me just show you, or give you a quick demo, of how this works. It starts with a collector. I have three collectors. I already showed you. One for each data center. What I’m going to do is install another collector locally. This will take a second to download. So I’ll jump back and we’ll talk through what’s happening here. This is automatically reaching out, grabbing our default collector. It’s downloading it. It’s installing. It will install on any system–whether it’s, you know, it’s a Linux system, Windows system, Mac. It is primarily
If we go back, let’s install. I can show you the sources that we have available. These are configured, but we also have all the available sources. This is what I’ve mentioned a few times, a hundred and fifty different sources that are available through BindPlane. We have the vast majority of AWS technologies, Google Cloud, Azure, but then we also have on-prem solutions. So things like, for compute, we have Dell EMC compute. We have Lenovo, Nutanix, HPE, Cisco, many others. We have databases, a wide range of enterprise and open-source databases. Load balancers, networks and storage.
Q. All of those can be configured on any of these collectors?
A. Anything that can be reached, that as network access to that collector.
Q. In the connector screen, I saw Cloud Formation?
A. Yes, we have a deployment template for Cloud Formation as well in the AWS collector, yes, exactly. It’s just a simplified version. If I was to click this, and I can choose a region that I want that to be deployed in, and then click a button, and it will automatically pull up a confirmation.
Q. What does it actually deploy? A load balancer for instance, so it’s it’s a single instance?
A. What we do is we allow you to bundle it, to group collectors together. So if you deployed two of those, you can go into the UI here, choose a collector and in this case, this is already a group, so I can just click “Add collector” and bundle as many of those together as I want. So I can put two, put five collectors inside of a group, and those now become high availability, and distributed load for all of your collectors.
Q. So each collector has a certain amount that it can handle?
A. You know, you can see most of these are single and they have dozens, if not hundreds, of different integrations installed. Once you get to a certain point, you start to distribute that into multiple.
Q. So you don’t go big you just scale out? Great.
A. Exactly, yeah.
Q. Or if you want, intend to, do that high availability, then you have two of those, right?
Q. Do you have a similar thing for Azure, or some other clouds as well?
A. It’s on the way, but it’s it’s not there yet.
Q. Currently, you just have AWS?
A. Yes, right, but Azure and GCP will be out shortly.You could also install, because you can run a Linux image on Azure. You can still install our collectors, yes. It’s about how folks install them on any platform. We’re just making it as easy as possible. So we’ve got the AWS CloudFormation template we have a, containerized–a Docker container that’ll be out shortly, and then GCP and Azure templates as well, soon. Both of them come from customer requests. That’s how we determine our roadmap for sure. So, you know, we get feedback all over the place for connectors, for cloud support, for integrations. That all feeds into this.
So this installation finished here. It is now running and it should show up. And here is a new collector. So there we go. So what happened was, I installed this collector. It’s it then called out to BindPlane and registered itself. And now it’s available to be used within my environment.
So I can now do something. Like if I want to add a source. Take a look at this. I have a Postgres database running on my laptop. So I will try to monitor this Postgres database. So I’m doing this off on the web UI. So what’s happening is, I build this configuration. As soon as I hit “test connection.”
When I installed that collector, it didn’t have any integrations, so it’s a collector with nothing installed. As soon as I hit “test connection,” it downloads the correct, the latest version of our Postgres integration. It then pushes this configuration down and will attempt to make a connection. So you know before we save it. I know that it now connects. It’s able to gather all the data and as soon as I hit “add.” This is now going to be available. It’s going to show up in our overview in just a second. There we go. So now we have a Postgres database running. I’m monitoring at my laptop. Now being sent to BindPlane. That data is going to both of those destinations that I have configured.
Q. So you can set policy and monitoring patterns from BindPlane and push out to the child?
A. Yes. In fact, that’s a huge part of the benefit of this. For each one of these, you know, you see and we have a hundred or so integrations here. In the past, that would mean that you have a hundred different agents installed somewhere. That you don’t have to manage. That you have to configure with a JSON file or whatever on the system. In this case, we can we can update them remotely. You can set automatic updates. So every time Blue Medora releases a new Postgres database, it’ll automatically download it and update the collectors will update on their own. All the configuration can be pushed as opposed to configured locally.
Q. Are there any plans to make BindPlane on site?
A. It’s something that we’ve seriously considered. And just to give you a little bit of the underpinnings of it, we’ve designed it to be possible to do that from the beginning. So we use a Kubernetes environment that allows it to port from one location to another. So certainly there’s been some interest.
Q. As I can see this be incredibly meaningful to address enterprises agent proliferation. Being able to solve that problem on site without having the use cloud would be a god send. Is their particular service mesh you guys you’re using?
A. This is actually all running on AWS for this component. And then just briefly to show you actually, where the data goes. You know, now I’m jumping into New Relic. This is just a standard New Relic console. Very quickly you see all the data that we have available. So all of those integrations are now available within New Relic. If I wanted to go down and look at let’s say, VMware virtual machines, I maybe will look at the guest OS. Alright so now I’ve just taken all the data that we’re monitoring we pushed it over to New Relic and I’m looking at the specific guest OS. This is just to show you a simple example. All the data that you have is now available in a new platform. We go further with these by showing, we actually build on top of this applications, with it what they call apps. This is just ways to visualize these components so this is one for Oracle. If we jump back I can show you that we had the Postgres database that we set up. Now you see this localhost Postgres database is available. So, the one I just did a minute ago is now sending data into New Relic.
Q. Is it consuming resources within New Relic or like the collector itself?
A. It’s really low there. These are not highly-resource intensive. I mean, that’s going to scale depending on how many integrations you have on it. But like I said usually we scale out to keep them relatively small. You’re talking about you know a couple of gigs of memory on something in low CPU usage.
Q. So I’ve seen a monitoring system kill applications.
A. That’s um. So, I think I was answering a different question. If you’re talking with the load that’s actually put on the application itself. So, if we’re looking at the load on a Postgres database we keep it minimal. Because we are we’re not on the box itself, it reduces the actual usage of system resources. We’re going remotely through API calls and, of course, it’s something that it’s critical in monitoring to watch what types of API calls we’re making–what the load is on those integrations or on those systems.
Q. How is this all licensed?
A. So this is a traditional SaaS license. It’s a subscription model. We do a model of a bucket of metrics that you can have sent. Then it’s discounted very heavily for additional platforms. So if you’re doing this for three platforms, you’re not paying three times the price.
Q. Is it sold through the partner channel or direct?
A. This is this is currently direct. Then we’re partnering with platforms to simplify that as much as possible.
Q. So with this be added to the VMware VCPP program, much like the management packs are for vROps? or is this is this intentionally separated out?
A. I can’t answer that. I can tell you that we’re moving towards that or trying to make it as simple as possible.