We are using it for monitoring our cloud environment and detecting misconfigurations in our hosted accounts in AWS or Azure.
We are using it for monitoring our cloud environment and detecting misconfigurations in our hosted accounts in AWS or Azure.
As the security operations team, our job is to monitor for misconfigurations and potential incidents in our environment. This solution does a good job of monitoring those for us and of alerting us to misconfigurations before they become potential security incidents or problems.
We've set the tool up so that it provides feedback directly to the teams responsible for their AWS or cloud accounts. It has been really helpful by getting information directly to the teams. They can see what the problem is and they can fix it without us having to go chase them down and tell them that they have a misconfiguration.
The solution secures the entire spectrum of compute options such as hosts and VMs, containers and Containers as a Service. We are not using the container piece as yet, but that is a functionality that we're looking forward to getting to use. Overall, it gives us fantastic visibility into the cloud environment.
Prisma Cloud also provides the data needed to pinpoint root cause and prevent an issue from occurring again. A lot of that has to do with the policies that are built into the solution and the documentation around those policies. The policy will tell the user what the misconfiguration is, as well as give them remediation steps to fix the misconfiguration. It speeds up our remediation efforts. In some of the cases, when my team, the security team, gets involved, we're not necessarily experts in AWS and wouldn't necessarily know how to remediate the issue that was identified. But because the instructions are included as part of the Prisma Cloud product, we can just cut and paste it and provide it to the team. And when the teams are addressing these directly, they also have access to those remediation instructions and can refer them to figure out what they need to do to remediate the issue and to speed up remediation on misconfigurations.
In some cases, these capabilities could be saving us hours in remediation work. In other cases, it may not really be of value to the team. For example, if an S3 bucket is public facing, they know how to fix that. But on some of the more complex issues or policies, it might otherwise take a lot more work for somebody to figure out what to do to fix the issue that was identified.
In terms of the solution’s ability to show issues as they are discovered during the build phases, I can only speak to post-deployment because we don't have it integrated earlier in the pipeline. But as far as post-deployment goes, we get notified just about immediately when something comes up that is misconfigured. And when that gets remediated, the alert goes away immediately in the tool. That makes it really easy in a shared platform like this, where we have shared responsibility between the team that's involved and my security operations team. It makes it really easy for us to be able to go into the tool and say, "There was an alert but that alert is now gone and that means that the issue has been resolved," and know we don't have to do any further research.
For the developers, it speeds up their ability to fix things. And for my team, it saves us a ton of time in not having to potentially investigate each one of those misconfigurations to see if it is still a misconfiguration or not, because it's closed out automatically once it has been remediated. On an average day, these abilities in the solution save my team two to three hours, due to the fact that Prisma Cloud is constantly updating the alerts and closing out any alerts that are no longer valid.
The policies that come prepackaged in the tool have been very valuable to us. They're accurate and they provide good guidance as to why the policy was created, as well as how to remediate anything that violates the policy.
The Inventory functionality, enabling us to identify all of the resources deployed into a single account in either AWS or Azure, or into Prisma Cloud as a whole, has been really useful for us.
And the investigate function that allows us to view the connections between different resources in the cloud is also very useful. It allows us to see the relationship traffic between different entities in our cloud environment.
The integration of the Compute function into the cloud monitoring function—because those are two different tools that are being combined together—could use some more work. It still feels a little bit disjointed.
Also, the permissions modeling around the tool is improving, but is still a little bit rough. The concept of having roles that certain users have to switch between, rather than have a single login that gives them visibility into all of the different pieces, is a little bit confusing for my users. It can take some time out of our day to try to explain to them what they need to do to get to the information they need.
I have been using Palo Alto Prisma Cloud for about a year and a half.
We really have had very few issues with the stability. It's been up, it's been working. We've had, maybe, two or three very minor interruptions of the service and our ability to log in to it. In each case there was a half an hour or an hour, at most, during which we were unable to get into it, and then it was resolved. There was usually information on it in the support portal including the reason for it and the expectation around when they would get it back up.
It seems to scale fine for us. We started out with 10 to 15 accounts in there and we're now up to over 200 accounts and, on our end, seemingly nothing has changed. It's as responsive as it's ever been. We just send off our logs. Everything seems to integrate properly with no complaints on our side.
We have nearly 600 users in the system, and they're broken out into two different levels. There are the full system administrators, like myself and my team and the security team that is responsible for our cloud environment as a whole. We have visibility across the entire environment. And then we have the development teams and they are really limited to accessing their specific accounts that are deployed into Prisma Cloud. They have full control over those accounts.
For our cloud environments, the adoption rate is pretty much 100 percent. A lot of that has to do with that automated deployment we created. A new account gets started and it is automatically added to the tool. All of the monitoring is configured and everything else is set up by default. You can't build a new cloud account in our environment without it getting added in. We have full coverage, and we intend to keep it that way.
Tech support has been very responsive. They are quick to respond to tickets and knowledgeable in their responses. Their turnaround time is usually 24 to 48 hours. It's very rare that we would open anything that would be considered a high-priority ticket or incident. Most of the stuff was lower priority and that turnaround was perfectly acceptable to us.
This is our first tool of this sort.
The initial setup was really straightforward. We then started using the provided APIs to do some automated integration between our cloud environment and Prisma Cloud. That has worked really well for us and has streamlined our deployment by a good deal. However, what we found was that the APIs were changing as we were doing our deployment. We started down the path we created with some of those integrations, and then there were undocumented changes to the APIs which broke our integrations. We then had to go back and fix those integrations.
What may have happened were improvements in the API on the backend and those interfered with what we had been doing. It meant that we had to go back and reconfigure that integration to make it work. My understanding from our team that was responsible for that is that the new integration works better than the old integration did. So the changes Palo Alto made were an improvement and made the environment better, but it was something of a surprise to us, without any obvious documentation or heads-up that that was going to change. That caught us a little bit out and broke the integration until we figured out what had changed and fixed it.
There is only a learning curve on the Compute piece, specifically, and understanding how to pivot between that and the rest of the tool, for users who have access to both. There's definitely a learning curve for that because it's not at all obvious when you get into the tool the first time. There is some documentation on that, but we put together our own internal documentation, which we've shared with the teams to give them more step-by-step instructions on what it is that they need to do to get to the information that they're looking for.
The full deployment took us roughly a month, including the initial deployment of rolling everything out, and then the extended deployment of building it to do automated deployments into new environments, so that every new environment gets added automatically.
Our implementation strategy was to pick up all of the accounts that we knew that we had to do manually, while we were working on building out that automation to speed up the onboarding of the new accounts that we were creating.
We did all of that on our own, just following the API documentation that they had provided. We had a technical manager from Palo Alto with whom we were working as we were doing the deployment, but the automated deployment work that we did was all on our own and all done internally.
At this point, we really don't have anybody dedicated to deployment because we've automated that process. That has vastly simplified our deployment. Maintenance-wise, as it is a SaaS platform, we don't really have anybody who works on it on a regular basis. It's really more ad hoc. If something is down, if we try to connect to it and if we can't get into the portal or whatever the case may be, then somebody will open a ticket with support to see what's going on.
We have seen ROI although it's a little hard to measure because we didn't have anything like this before.
The biggest areas of ROI that we've seen with it have been the uptake by the organization, the ease of deploying the tool—especially since we got that full automation piece created and taken care of—as well as the visibility and the speed at which somebody can start using the tool. I generally give employees about an hour or two of training on the tool and then turn them loose on it, and they're capable of working out of it and getting most of the value. There are some things that take more time to get up to speed on, but for the most part, they're able to get up to speed pretty quickly, which is great.
The pricing and the licensing are both very fair.
There aren't any costs in addition to the standard licensing fees, at this time. My understanding is that at the beginning of 2021 they're not necessarily changing the licensing model, but they're changing how some of the new additions to the tool are going to be licensed, and that those would be an additional cost beyond what we're paying now.
The biggest advice I would give in terms of costs would be to try to understand what the growth is going to look like. That's really been our biggest struggle, that we don't have an idea of what our future growth is going to be on the platform. We go from X number of licenses to Y number of licenses without a plan on how we're going to get from A to B, and a lot of that comes as a bit of a surprise. It can make budgeting a real challenge for it. If an organization knows what it has in place, or can get an idea of what its growth is going to look like, that would really help with the budgeting piece.
We had looked at a number of other tools. I can't tell you off the top of my head what we had looked at, but Prisma Cloud was the tool that we had always decided that we wanted to have. This was the one that we felt would give us the best coverage and the best solution, and I feel that we were correct on that.
The big pro with Prisma Cloud was that we felt it gave us better visibility into the environment and into the connections between entities in the cloud. That visualization piece is fantastic in this tool. We felt like that wasn't really there in some of the other tools.
Some of the other tools had a little bit better or broader policy base, when we were initially looking at them. I have a feeling that at this point, with the rate that Palo Alto is releasing new policies and putting them into production, that it is probably at parity now. But there was a feeling, at the time, among some of the other members of the team that Palo Alto came up short and didn't have as many policies as some of the other tools that we were looking at.
I would highly recommend automating the process of deploying it. That has made just a huge improvement on the uptake of the tool in our environment and in the ease of integration. There's work involved in getting that done, but if we were trying to do this manually, we would never be able to keep up with the rate that we've been growing our environment.
The biggest lesson I've learned in using this solution is that we were absolutely right that we needed a tool like this in our environment to keep track of our AWS environment. It has identified a number of misconfigurations and it has allowed us to answer a lot of questions about those misconfigurations that would have taken significantly more time to answer if we were trying to do so using native AWS tools.
The tool has an auto-remediation functionality that is attractive to us. It is something that we've discussed, but we're not really comfortable in using it. It would be really useful to be able to auto-remediate security misconfigurations. For example, if somebody were to open something up that should be closed, and that violated one of our policies, we could have Prisma Cloud automatically close that. That would give us better control over the environment without having to have anybody manually remediate some of the issues.
Prisma Cloud also secures the entire development lifecycle from build to deploy to run. We could integrate it closer into our CI/CD pipeline. We just haven't gone down that path at this point. We will be doing that with the Compute functionality and some of the teams are already doing that. The functionality is there but we're just not taking advantage of it. The reason we're not doing so is that it's not how we initially built the tool out. Some of the teams have an interest in doing that and other teams do not. It's up to the individual teams as to whether or not it provides them value to do that sort of an integration.
As for the solution's alerts, we have them identified at different severities, but we do not filter them based on that. We use those as a way of prioritizing things for the teams, to let them know that if it's "high" they need to meet the SLA tied to that, and similarly if it's "medium" or "low." We handle it that way rather than using the filtering. The way we do it does help our teams understand what situations are most critical. We went through all of the policies that we have enabled and set our priority levels on them and categorized them in the way that we think that they needed to be categorized. The idea is that the alerts get to the teams at the right priority so that they know what priority they need to assign to remediating any issues that they have in their environment.
I would rate the solution an eight out of 10. The counts against it would be that the Compute integration still seems to need a little bit of work, as though it's working its way through things. And some of the other administrative pieces can be a little bit difficult. But the visibility is great and I'm pretty happy with everything else.