What is a CMDB?

(Host: Michelle Dawn Mooney)

Welcome to the Ultimate Guide to the Device 42 Configuration Management Database, or CMDB, podcast series. I’m your host, Michelle Dawn Mooney. And in this episode, we are going to explore what a CMDB is, the elements of a CMDB, including attribute relationships, and why a clean update to date CMDB is important for IT operations. And joining us today from Device 42 is Chris Benini, who is Director of Solutions Architects, and James McFadden, Senior Solutions Architect. Thank you so much for being with me today. Looking forward to this conversation. So welcome. 

Yes, thank you for having us today. 

(Host: Michelle Dawn Mooney)

Okay, so let’s start off here. What exactly is a configuration management database? And why is it important for IT operations? 

(Guest: James McFadden)

So it’s a place for all the things like so when you start talking about an organization and how they manage all of their assets, either physical, virtual, business functions, things like that, you have to have a central repository where all that information is located. But in order for you to be effective and efficient with your organization, if you don’t know where everything is, or you don’t have information about all your assets that you have readily available to you, you can’t effectively do your job. So in layman’s terms, it’s a giant book of all the things. It’s a list of everything about everything about your organization. 

(Guest: Chris Benigni)

Chris, you want to? Yeah, no, I concur there. And you’re probably gonna be hearing me say I agree a lot, just on some of the terminology around here. Yeah, I mean, it really is a comprehensive view of all your, you know, your assets. And it’s not just your IT estate. I mean, it could be, you know, hardware, software, it could be system data, it could be like business metadata, it could be personnel, just tracking who the contact owner is of an application. You know, so, you know, I think those of you that have a CMDB today, and you know, as we go through some of this, some of the webinar, you know, you’ll see what’s in how the importance of a CMDB, but also the driver behind it, too, is really important. You know, just having a CMDB, having a place where data is aggregated, where you can store that information is really important, because it does help with things like meantime to resolution and just being able to, you know, help with any initiatives, whether it’s a hardware refresh, whether it’s an acquisition that you’re going through, the idea is to make sure, you know, that you have visibility into all your infrastructure, and anything that you know, that participates with your infrastructure as far as, you know, personnel and, you know, other types of metadata that I mentioned. I think the next piece that we wanted to talk a little bit about is what is the goal of a CMDB. And I think the keen thing there, and I like using this terminology, is really a source of truth. And I think it’s important to know that, and James, you and I talked a little bit about this, but a CMDB is really not a ticketing system. It is, you know, not a discovery tool. CMDB is really a configuration management database. So it’s a place where you’re tracking CIs, you’re tracking relationships to those CIs. And I think, you know, one of the things is just really having visibility into it. So a source of truth is really the ability to say, is this a place where everybody can go to no matter what department and get information about what they see in their environment, you know, whether it’s customer data, whether it’s internal data. And I think one of the things, just kind of thinking out loud, James, is, you know, what are your thoughts a little bit around the goal of a CMDB on top of what I had mentioned? 

(Guest: James McFadden)

So I think really a fundamental goal is enabling your business to function in an effective manner. I touched on that previously, but it’s really heavily used in ITIL practices, for example. So say you’re trying to move an organization forward and you function in that ITIL practice, right? That’s how you actually proceed and continuously develop your infrastructure, your development process, your business goals. You can’t do that without an accurate CMDB. I mean, shameless plug here, you know, we work for Device42 and CMDB is a portion of our product. It’s a big part of our product. So having a lot of accurate information within our product is really heavily relied on. 95% of the customers that actually come to Device42 are looking for that central repository, looking for that single source of truth that you mentioned that actually has all the information about all of their IT assets, all of their devices, whether they physical or virtual, even business functions. So business services, for example, is a really, really good example of things that can go into a CMDB that’ll help move your business forward. 

(Guest: Chris Benigni)

Yeah. And one last comment too is as you see more and more customers moving to a hybrid estate or a multi-cloud environment, there’s a lot more functions that you have to track. So it’s not just an IP address, there’s tons of different services that are out there. So think of a platform as a service. So how do you aggregate all this? How do you make sense of it? And I think obviously you have the CLI platforms for all these different vendors, but if you’re using Google Cloud, if you’re using AWS, if you’re using Azure, you have on-prem infrastructure. I mean, think about, you have applications that are made up of on-premise and cloud infrastructure. How do you properly store all that? And I think it’s important to know that as things are becoming more modernized, things are becoming more infrastructure, that it becomes more and more apparent that having a CMDB is important for any organization. 

(Host: Michelle Dawn Mooney)

Okay. So now we’ve established what is a CMDB, why it’s so important, but let’s dig a little deeper. What are the elements of a CMDB? 

(Guest: Chris Benigni)

Yeah. So great question. So, you know, I think this is a good thing to start to also for folks to kind of analyze and understand, you know, do they have a CMDB? What’s being presented to them as a CMDB? So CMDB is going to be a list of configuration items. So I think I mentioned things around software and hardware, you know, business metadata, personnel information, different services. You know, if you have a business service catalog or application catalog, you know, whether these are applications that you have SLAs to meet and making sure that they’re up and running, you know, or these are internal applications or test environments. These are all things that become what we call configuration items or CIs. And on top of that, you’ll have all these different attributes, right? So for each of these components, they’ll have metadata that’s about them. So that’s, you know, being able to say that there’s a device that has an operating system, it has an IP address, it has a hardware address, it has a subnet that IP address belongs to. So these are all what you call configuration items or CIs, and then these are all the attributes that are part of them. I think what’s important though, is as you’re looking at some of these CMDBs to make sure that how much of this is pre-configured, right? So if you’re looking into a platform today, how much do you have to build out of the box? And I think what’s really important about CMDBs is also the flexibility around them, right? So being able to, not everything out of the box is going to have, you know, a core code or something that is specific to your organization. So you want to make sure that you have the ability to add custom field attributes. And all this falls into a CMDB. So being able to have some line of metadata that can track common inventory data that you have today, as well as a lot of customization data that might be specific to your organization. Now on those CIs, there’s also relationships and I’ll kind of, James, I’ll pivot that to you just on how important having an attribute relationship is to a CMDB. 

(Guest: James McFadden)

So when you, when we’re talking about CMDB in a place that houses all the things, all of your CIs, right, understanding how those things relate to one another is super key. So like you were mentioning, just listing off the different types of CIs and configurable configuration items that are in a CMDB. One of the things you mentioned actually was like, hey, what subnet does this belong to, right? Well, there’s going to be a lot of other things that belong to that subnet too, including infrastructure that makes that subnet function all together. So switch fabric, routers, firewalls, load balancers, things like that, right? So when we’re starting to talk about what’s important to actually have in a CMDB repository, not only having the individual items, but having the, hey, this device relies on that device or this system or application relies on that system or application. So that when you start leveraging this CMDB in say an ITSM platform, so say you’re feeding all of this information into something like Atlassian’s JSM, for example, you can start to leverage that information to make your workflows more effective. So having a CMDB where you can say, hey, this device relates to this other device, you get an alert on an alerting system that also feeds into the ITSM platform and says, hey, a specific device is having a problem. You can quickly and effectively say, hey, that’s going to affect this and this and this and this and this, specifically because you have already discovered all of those attribute relationships. So knowing what relies on what and what systems talk to what systems, hugely beneficial for a CMDB. And then that actually kind of gets into making sure that those relationships are actually kept up to date. So it’s one thing to have information about CIs and keep that information up to date, but also having the relationships up to date. And Chris, I don’t know, you want to talk a little bit about updating in general and making sure stuff is accurate and how important that is. 

(Guest: Chris Benigni)

Sure. Yeah. And that’s an important piece that you bring up, James. I have my Device 42 shirt on here, so I’m going to sponsor Device42 in this conversation. But I will tell you one of the big things, so if you have a CMDB right now, it doesn’t matter how up to date it is or whatnot, it’s better than no CMDB. So you have a place you can go to. Now, when you’re thinking about a CMDB, we’ll touch a little bit about what really is, what kills a CMDB or what is a bad CMDB. But ideally, you do want something that’s keeping that data up to date. So a lot of customers that we speak to today, a lot of them are doing this manually, even when there’s solutions out there like an ITSM tool that has a discovery solution, some customers aren’t using it. So what they’re doing is they’re manually updating their CMDB. Now that’s good to some degree, but you’re only human, you’re going to make mistakes, you’re going to forget to update things, you’re also not going to be updating everything. So it’s how important it is to keep that data up to date, because when you are looking at an incident or a change request, and all of a sudden, something goes down, if you haven’t updated a device or its relationships or what application it supports, or you forget to update that there’s a new contact owner or app owner, sometimes you spend 80% of, a company will spend 80% of their time trying to identify who to contact before they can even remediate an issue. So it’s really important to make sure that a CMDB is up to date. And a lot of times that really leads to having a discovery tool or something in place, whether you’re scripting APIs, or whatever, whatever the case is in connecting to different devices. But ideally, you want to look at something like a discovery tool that will collect that data, and then feed that into a CMDB and help enrich that CMDB and keep it up to date. 

(Host: Michelle Dawn Mooney)

Okay, so we’ve established a lot of good things that can come from CMDBs and why they’re so important. But let’s take a look on the other side for a moment. What kills a CMDB? 

(Guest: James McFadden)

Yeah, there’s a lot of different things that can make a CMDB I don’t even, I don’t even want to say it makes it a bad CMDB so much as makes it non functional. So when we start talking about things that are bad for a CMDB, obviously bad data, bad sources for information, right? So if you’re feeding a CMDB data through some integration from some system, it’s out of date, your effectiveness is only going to be as good as the data is involved, right? So when we start talking about not having accurate data, not tracking changes about your CIs, not being able to possibly access the data that you need due to a number of restrictions, whether that be security protocols, and security is a good thing, right? But having the individuals who are able to resource by having the individual individuals that are able to access the necessary resources is going to be huge. And then lack of standardization and getting too many hands in the pot with their own opinions about how something should be structured, and not having some kind of organizational structure for that CMDB, CMDB is definitely going to be problematic. I know, Chris, you’ve seen some results of bad CMDBs in the past, like, what does that look like from a business perspective, when you have bad data, inaccurate data, improperly tracked everything. 

(Guest: Chris Benigni)

So there’s a couple things that come to mind, I mean, you know, one thing is cost, like, if you think about it, cost can be a lot of things, cost can be, you know, having to have pay employees overtime and longer hours, because they can’t resolve issues quickly, because they just don’t have a CMDB or source of truth to go to, to help remediate issues. Another thing could be, you know, downtime, I mean, if you’re, if you’re an application, if you host an application, and, you know, you’re, you have a portal, and, you know, maybe your, your sweet time is around Christmas, or, or Black Friday, and all of a sudden, your system’s down, I mean, think about the money that you that that you lose during that period of time. And whether even if it’s just for an hour, we’ve seen some crazy numbers with customers, but it costs is a big thing, you know, that and it can be it can have a lot of variances as far as supporting the CMDB, or, you know, just just downtime from not having an accurate CMDB. One other thing is around increasing mean time to resolution. So if you have the whole goal is to minimize MCTR, and, you know, and I think it’s, it’s, it’s a goal across any organization, doesn’t matter what vertical, and really, a big, not common denominator there is, is not having a source of truth or an accurate CMDB. So having a poor, you know, bad CMDB, for many reasons, as James was mentioning, can lead to, you know, an increased time meantime to resolution. And like I said, that all relates back to cost. So there’s a lot of there’s, there’s a lot of variances from that. The other thing too, is planning, right? So you can have poor decision making on the confidence level of making a decision on you know, should we take the server down? Who do we, you know, do we know the full impact of this? You know, these are all things, or maybe planning, you know, so a lot of times what I find in an organization today, everybody’s in a reactive state. And it’s usually a result of, you know, not having the right tooling in place, so that they can utilize their time in different areas of the business, and actually do their job properly and be more proactive. So a lot of times it’s reacting to things. So I think that’s a big piece there is, is really, you have poor decision making when you don’t have accurate data, and it leads to bad results. And then lastly, a big thing I kind of hinted on it a little bit, was around maximizing personnel skill set. I mean, you could have, you can hire one of the smartest network engineers and say, this guy is going to help me build out my network. If he’s busy troubleshooting issues and spending so much time doing that, I mean, you’re, you’re not leveraging the skill sets that you paid this person for. You’re not helping them grow. You’re not allowing the organization to grow. You know, so there’s a lot of things that help that this prohibits when you don’t have a source of truth. And again, I think it all really goes, a lot of these things, there’s a lot of overlap, but it all comes down to, you know, cost and mean time to resolution, poor decision making, you know, maximizing personnel skill sets, and there’s, there’s obviously a lot more, but these are big components, I think of, you know, what are the results of really having a bad CMDB? 

(Host: Michelle Dawn Mooney)

So we’ve covered a lot of ground here. But what would you say are some key takeaways that you would like our audience to leave with today? 

(Guest: Chris Benigni)

Sure. James, do you want to start? 

(Guest: James McFadden)

Yeah, sure. So number one, if you’re trying to be effective in your organization, and you’ve outgrown spreadsheets, you need a CMDB. So what is your CMDB? What do you currently have? Is it actually functional for you? What’s driving the data in your CMDB? Is it actual, accurate, empirical information? Is it intervened by human fingers on a keyboard? Is it something that’s actually programmatic and effective, or is it garbage data that’s going to wind up sending you the wrong direction? Definitely a couple big things to do, a couple of big things to work on, look out for there, you know, accuracy of your data is going to be huge. And then having a CMDB that’s actually functional for your organization, specifically, is going to be a big deal. Chris, do you want to tackle? 

(Guest: Chris Benigni)

Yeah, you know, I think one thing that I’ve noticed too, is obviously, ITSM is a big, is a big area. It’s a big, it’s a big thing that a lot of customers already have. And in ITSM tools, for the most part, have a CMDB. And you know, and I think what I see some of these tools lacking is really, like advanced enterprise discovery, you know, and that that is a way of how do we efficiently scan data across the network, whether it’s a device, whether it’s software installed on device, and help enrich my CMDB. So one thing I just want to mention is, you know, don’t, you know, don’t always reside on just the CMDB that you’re kind of stuck with, with the ITSM tool. I mean, in some cases, obviously, they might have a discovery tool that can feed to that. But you know, there’s also this notion of best in breed. And you know, where I’ll kind of surface this is a little bit is device 42 is an agentless discovery solution. We’re sort of an advanced enterprise discovery solution. And we integrate with multiple CMDBs. And it’s primarily for that purpose that it is that best of breed, right? So best of breed really provides better functionality. It allows organizations to be flexible and agile. So if you have, you know, let’s say a different initiative, something all of a sudden switches or there’s an acquisition, how do you really pivot when you’re dedicated to one single vendor? And I think sometimes, you know, again, going back to having, you know, a discovery tool that’s really feeding into a CMDB, you know, is going to help improve a lot of things like mean time to resolution, it’s going to improve efficiency. So I think it’s really important, you know, as you’re looking at what you have today, don’t be discouraged. You know, should I not be looking at something else? Or how can I do this better? And also, you know, think about, you know, some of these worst case scenarios, where, you know, what if this goes down? Or what if this went down? And then just try to think about how you would solve that today? And if you feel that, you know, you’re going through hoops, and there’s no accurate way, you’re not confident about that decision, then take a look at, you know, how do I find data that can, or how do I find tool sets that can feed into an existing CMDB? Or how do I look for what tools out there offer CMDB? But you know, those are some things that just to keep in mind, you know, again, I’m tagging on on top of what James said here is, I think those are important things to keep in mind. You know, even if you have something today, how good are you at being able to identify if there was an issue, you know, can, you know, can you resolve that? 

(Host: Michelle Dawn Mooney)

Chris Benigni, Director of Solutions Architects and James McFadden, Senior Solutions Architect for Device 42. Thank you so much for being here today. Great conversation, and appreciate your time today. So thank you. And I want to thank all of you for joining us today for the Hitchhiker’s Guide to Device42 CMDB. And stay tuned for our next episode on CMDB challenges and best practices. I’m your host, Michelle Dawn Mooney. Thanks again for joining us. We hope to connect with you on another podcast soon.