Cisco Defense Orchestrator Review

Provides visibility into entire infrastructure and bulk changes save time and resources

What is our primary use case?

Most of the time we use it for the simplicity, for streamlining security policy management. We have other layers of stuff that we use with Cisco, from an integrated standpoint. Defense Orchestrator brings everything together.

How has it helped my organization?

For one particular client, we had almost a 20 percent remediation on some of their equipment as a result of all kinds of attacks from the desktop department. We got them down to a zero percent remediation. In other words, in retrospect, their data center and their desktop division went to zero percent when we deployed everything along with Defense Orchestrator. It was a huge success for the client. Defense Orchestrator was instrumental in that. In terms of visibility and getting everybody involved, it was simple, scalable, and saved them tons of time, which in turn saved them money. Sadly enough, they didn't need as many people any longer in certain departments. They were able to move them over, get them training and move them out. They got more projects done and had to do less firefighting. The biggest thing was that it allowed them to dial in, quickly, on what the threat landscape was for their architecture.

When it comes to making bulk changes across common tasks, like policy management and image upgrades, one of the biggest complaints that I had from a lot of network engineers, was that everything was GUI, that Cisco had gone to GUI. But they can do bulk changes on the CLI. That was a big win for them, being able to do that across all the ASAs without having to log into every single ASA and make changes. They can do a lot of bulk changes on the fly. It's a huge time-saver. The biggest benefit is obviously from the security standpoint, but at the C-level what they see are the cost savings. It's less billable time and fewer resources.

One of the biggest problems we were able to solve was due to its ability to use third-party apps, using a RESTful API and being able to integrate Splunk - things the clients already had in place - without any issues. That part was very easy. 

There's a lot of built-in stuff. You pull logs on the fly and you can troubleshoot problems when they come up, as well. That's been really helpful. It has solved clients pain points. 

When there are issues when they roll out configs, CDO allows us to do rollbacks really easily on a bulk level. That works really well too. It keeps track of "good configs."

In terms of simplifying security policy management across an extended network, if a lot of people are working on the same stuff, then the architecture has been broken up to different areas. Now, from a management standpoint, it is no longer a nightmare when I go in there and try to determine what is going on in the network. I have one "throat to choke." When I login, I have visibility into what is going on over the entire infrastructure. In case somebody left the door open, I have that visibility now.

Its effect on firewall builds and daily management of firewalls is that it's super-simple on new deployments. We haven't done any really large ones, but I've read some deployments where people have done thousands of ASAs with one massive import and there wasn't any downtime with respect to changing out equipment which was no longer under Smart Net.

Also, when we're looking at policies, it identifies the shadow rules. It notifies us about anything that will supersede other rules.

What is most valuable?

The simplicity, efficiency, and effectiveness of it are valuable.

There are a lot of templates that are already built-in. They give you quick-to-create and quick-to-apply policies that are typically a little more complicated for people.

What needs improvement?

Some of the issues we've had aren't really a CDO problem. For example, we had some MX devices that were blocking Windows Update from happening. We found out it was a Meraki issue, but it would have been nice if it had been flagged for us: "Hey, these updates are failing because the MX is blocking it." It wasn't a huge problem, but there was a loss of our time as well as the fact that the updates didn't get pushed out. You could look at that as a security issue but, at the same time, when updates won't run for any reason on certain machines, you freak out a little bit.

We thought it was a licensing issue with Microsoft or it could have been Dell EMC. But we were wasting time making all these phone calls and having people remotely troubleshoot it. The troubleshooters were saying, "Man, this looks like a network issue." They tethered a phone and joined it to the wireless on the phone to see if it would update and, boom, it started working. The weird thing was that when we switched it back over to the network, the Meraki was letting it through at that point. It would have been nice if CDO had let us know that that was an issue.

There are probably some things that it could do as far as some of the analytics are concerned, things I know it would be capable of: "Hey, why are all these requests coming in? The reason is that a firmware update needs to happen on the Meraki. It's a known issue." That would be helpful.

For how long have I used the solution?

We got some training from Cisco and it was in the fall of last year when we got really heavy into it, about eight months ago. There was some earlier development stuff through which we got some exposure to it. We're a Cisco partner, and our typical vendor of choice is either Tech Data or Ingram Micro and that's how we got some early exposure to it.

What do I think about the stability of the solution?

We don't have a large window of time to look back on, we don't have years of experience, but so far the stability has been pretty darn good compared to anything we've ever had.

What do I think about the scalability of the solution?

When it comes to scalability it's flexible, absolutely.

The largest network deployment that I've been involved with - we're not a very large company - had about 10 ASAs on the data center side and there were 29 other locations. There were less than 50, as far as the firewall devices go. At the largest deployment, the user count is somewhere a little over 1,000.

Scalability isn't an issue. We had some opportunities we didn't close, a university campus where the deployments were for about 15,000. We scoped it and scaled it out. The licensing gets a little different on some of the products when you go over 10,000 users. Sometimes the product line changes too in terms of design and scope.

How are customer service and technical support?

When we had to use tech support on the first setup, it was more for asking questions because we got pretty good training prior.

If you previously used a different solution, which one did you use and why did you switch?

There's a lot of different stuff, solutions which integrate into companies' ticketing systems. It depends on what your needs are. Even stand-alone, with FirePOWER, Umbrella, and AMP for Endpoints, there is Threat Grid - think CDO but on a very small scale. Prior to CDO, Cisco had, and they still have, Threat Grid. To me, Defense Orchestrator is a higher-scale evolution of Threat Grid. People wanted more, and that "more" was delivered with Defense Orchestrator. 

Threat Grid is like a small, short-line railroad; it handles a small area of traffic. In the metaphor it might take stuff off ships and put it on the back of 18-wheelers. CDO is more like a Class I railroad like Union Pacific or BNSF or Norfolk Southern. They're going to go all over the place, on a much larger scale. The strength and power that CDO has is huge. It's like comparing a lawnmower engine to a V12 from a Bentley or an Aston Martin.

There's a huge difference in cost between these solutions. With the smaller solutions there's lag, even if it's not huge. What you're getting for almost no cost is a huge, valuable piece. But it's not going to be the same type of visibility and logging speed that you're going to get with CDO.

What about the implementation team?

On our end, the initial setup was pretty straightforward. We did receive some training along the way. I had done some test deployments, which I would tell anybody to do. There are certain things you can do inside of Cisco's dCloud to prepare you for deployments. But overall, it's efficient, simple, and there's the visibility on the security side. Deployment is fast. As a security person, I love the visibility and the ease of use when doing my upgrades.

In terms of implementation strategy, even before making the sell, we start from that standpoint. I don't want to say we're in the tank for Cisco, but it's what we have bought into. A lot of our engineers have training in it and they get ongoing training. Maybe it would be different if Fortinet gave us a ton of more training on their stuff.

There's a community that we're connected with, so when there are issues we typically hear about them in the communities beforehand. We know limitations going into projects and already have a good idea, a vision, of what direction we're going to go, so we can start planning properly. Cisco does a good job of training us in that process: When we model and design everything, how it's going to be set up; and once everything is deployed, how to analyze, how to remediate, how to get visibility into everything.

The time to deploy depends on the size of the deployment. In my experience, the longest part of the process is getting everything built and approved in CCW (Cisco Commerce Workspace). From a deployment standpoint, it's pretty easy at our end. We get the equipment in place, we stage the gear, we have all the existing stuff in place from the original infrastructure. The longest we've taken is about a month, once we have the gear in place and all the configs lined out. Typically, we've done a lot of front-load work in the process. In general, from first meeting the client to completing the end-to-end process, we're in and out in 120 days.

Everybody has a different part that they play, including the people who are doing racking and stacking. For the size of deployment that we've been discussing, we typically need ten to 12 people. There can be some travel involved so sometimes we need resources elsewhere, depending on the scale of the client and how far they are spread out.

Once it's deployed, an example of maintenance requirements is a location with a 24-hour operation, three eight-hour shifts, meaning three people are monitoring it and it works fine. You might need a fourth if you include like a "float guy" for when people go on vacation or get sick.

What was our ROI?

Once up and running, we see value from it right away. The impact is immediate. The biggest problem I have now is that something that gets forgotten is how bad things were before the implementation. C-level people tend to forget that.

The biggest part of ROI is the improvement to the operations. Our clients with CDO are having fewer issues. Things are just not going down. People are more productive. I don't know if any of the organizations that I've been with have done a study, but from an IT ticketing standpoint, tickets are down to one-tenth of what they were. People are able to bring in new projects and think about new things. From a staff being overtaxed due to remediation, because so-and-so clicked on an email or there was an issue with some type of a compromise, now it's eerily quiet.

What's my experience with pricing, setup cost, and licensing?

If I had to say anything negative it's the price point. Clients who can't invest in the complete package, it's a disservice to them because they don't have everything. They don't have as many layers. They don't have Defense Orchestrator. It shortchanges the product. Going back to old school theory, you broke up your infrastructure so you weren't tied into one architecture, but that's not necessarily the case anymore. Even if you have other hardware, with APIs, a lot of Cisco stuff and gear integrates very well, even with other devices.

I'm more on the engineering side, I'm not in CCW (Cisco Commerce Workspace) as much as the sales team and the account managers are. But I can tell you that it's not inexpensive. But to be honest, there are not a whole lot of products that give you all those features. There isn't an apples and oranges comparison. You can't compare a McLaren or a Ferrari or a Lamborghini to a Smart Car. There are different purposes and different requirements. Typically, you're buying these devices because you want performance and you're willing to go the extra mile for whatever it is you're trying to protect, whatever your crown jewels are. Whereas with the other devices, in my opinion, people are just trying to save money and do a "best-effort" against some of these things.

If it were me and it was my company, and my main goal was to protect my infrastructure, then I'd be using Cisco devices.

There are all kinds of different costs and now there is the advent of Cisco DNA. Cisco DNA is where they have that service-as-a-service type of billing. There's a monthly cost that's tied in to give you some additional analytics and visibility into what's going on in your environment. It's like taking a little piece of Meraki, all the cloud analytics that are coming in from their cloud-control devices. It's that middle-of-the-road step from them with Catalyst switches. I haven't seen anything on the Fabric side, from a storage standpoint, but I think it's just a matter of time. You're going to be getting data on a different layer, analytics on everything.

What other advice do I have?

As an engineer, I would say that if you can afford it, you will not be sorry that you invested in it. There's no question of whether it's going to deliver. The question is more from a value standpoint, the size of your business. If you're a national company with multiple locations across the US, CDO is the direction you need to be going in. If you're a small company, 50 people or less, you can probably get by using Threat Grid. Medium-size businesses will probably also be okay with using something like that. From an outside-of-Cisco vantage point, for small and medium-size business, Fortinet does a pretty decent job. But when you start getting into large-scale enterprise, there isn't anything right now that's doing the things that CDO is doing to enable you to integrate.

Cisco still has Tetration. To me, they are giving me a taste of Tetration, which is very high-scale leveraging. Think CDO but well beyond that. It's a multi-million-dollar device, a 42U-rack equipment storage device which is going to manage any and all network transactions happening on any of my networks. Tetration is for Exxon or Apple or Google-type visibility into the infrastructure. CDO gives me a taste of that without spending millions of dollars.

The biggest lesson I've learned from using it is that it sure is nice when people buy it. It just makes our job a lot easier. If you ask me to get a job done, with CDO you're giving me all of the components that I need to make everything you're asking me to do a success.

When it comes to its security features around storing firewall configurations in the cloud, there are things about that I probably don't fully understand, from a security standpoint. We've been doing that kind of thing for a long time, so I'm confident in it. But I'm a security guy, so I don't really trust anything. But that's where everything's going. It's good to know that I've got backups. "Cloud" is such an overused word too. As long as you thought through the security of everything, it's just some other place. Your attack spectrum is everywhere nowadays. To me, the biggest security problem is the human element. When you start looking at it like that, the fact that it's stored in the cloud is not really that big a deal.

It's just a different way of doing business. These are things that traditionally, ten years ago, even five years ago, people weren't comfortable doing. Cisco was kind of late to the party in a lot of these things, but over the last three years - the acquisitions, the overall way they've attacked everything - they're doing the best job of bringing everything in.

There are all the products which they have through acquisition, such as the OpenDNS acquisition for Umbrella, and CloudLock is going to be integrated into that as well; the next-gen firewall of FirePOWER and that's the evolution going into the FTD. They made a lot of improvements with ISE, even though there were some complexities that caused a lot of my higher-end clients to frown. It seems like they've righted the ship on all those things. So, there's a lot of good things happening. There are more things that I'm not really talking about, such as the evolution of even their switches, going with the FTD architecture of using Lina - Linux ASA - to do a lot of those pieces. One thing that they still have to rethink is how they're going to integrate a lot of the stuff that's on the ASA alone with AnyConnect, into FTD and those types of devices. We've been very pleased with the overall experience.

In terms of the solution’s support for ASA, FTD, and Meraki MX devices, we have tons of clients who use all these devices. Since 2007, we've done over 2,000 medical facilities in the southeast Texas market, just using Cisco ASA firewalls. But in many cases, these places aren't large enough to use Defense Orchestrator. Now, if we took over complete management, I see how we could integrate CDO from an industry standpoint because a lot of these places are very similar. They use the same EMR practice management. They operate the same way on their infrastructure, have the same type of buildings. In many cases they're in the same building, a medical center. But they don't operate that way. They have independent practice managers. They're typically somewhere between 25 to 60 users. It would be nice to be able to have something like that. Maybe somebody really forward-thinking in my organization could possibly sell that idea, although I'm sure our legal department would tell us it's a bad idea.

When you start dealing with HIPAA, there's a whole lot more to it than just IT. In managing that side of things, we do a lot of compliance and testing. We give them a HIPAA compliance report from an IT standpoint. And a lot of that is difficult because it has to be answered by someone within the organization who is familiar with their processes; for example, how they're turning their screen in an encounter with a patient. To have something like Defense Orchestrator, where I could manage hundreds of clients - their ASA or their Meraki MX or FTD - that would be huge.

As for increasing our usage of CDO, we don't have it in our internal infrastructure yet, due to cost and the fact that our needs aren't that great. If we start doing some private cloud hosting or the like, I could see us utilizing it. That's one of our goals. We've got four data center locations where we're planning on rolling out Cisco UCS with some redundancy and failover. We're looking at CDO as our main point of visibility.

I would rate Defense Orchestrator a ten. The only caveat I have for anybody trying to decide on it would be in terms of the budget and does it make sense for you. Do you need a 10,000-pound hammer to drive it home? We have a wide variety of clients in terms of size. Most people are somewhere in the middle-to-upper echelon with us. Others, and this is going to sound ugly, can't afford to use our services, because they're just looking for break-fix IT. They're still doing things the old-school way. Half of their data is compromised. They've been through several ransomware and malware attacks to the point where it has crippled their businesses. I don't know how those people operate.

It's difficult because the attack spectrum is in our backyard. As a security guy, with the things that are being done and that happen, I just don't know how people do it. That's especially true if they're using a static firewall or if they have in-house equipment and services opened up to the public. If they're using a static firewall and trying to do traditional things like port-forwarding, we see that. We walk in there and they're saying, "Everything's running really slowly." And they're completely compromised. We had somebody who couldn't place phone calls. Somehow, half their trunks had been compromised and were being used for a telemarketing service in Philippines. It's to the point where, if you're a fireman, you just let it burn. They need insurance at that point because they have massive problems.

Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor. The reviewer's company has a business relationship with this vendor other than being a customer: Partner.
1 visitor found this review helpful
Add a Comment
Sign Up with Email