What is our primary use case?
We used it only for six months. Initially, it turned out to be a good product, but then we had an issue, so we stopped using it. We are now using CrowdStrike.
From an endpoint perspective, we have a heterogeneous environment. We have Windows, we have Mac, and we have Linux endpoints. We deployed it on all the endpoints, all different operating systems, and cloud instances as well. Our AD was also integrated along with the identity solution, but the issues specifically get reported on the endpoints for open-source or Linux. That is why we decided not to move forward with it.
By implementing SentinelOne Singularity Complete, we wanted security for our endpoints. After COVID, endpoint security became even more critical because our perimeter was more exposed. It was expanding wherever the end users were, so endpoint security became much more critical. Previously, in terms of endpoint security, the traditional antivirus, anti-malware, and endpoint protection were disconnected systems. We did not have any offline correlation, log collection, or policy management, whereas SentinelOne, as well as CrowdStrike, come with a central console. For compliance requirements, such as ISO, SOC 2, or PCI, we have to provide evidence in terms of the status of the endpoint patches and security posture. That is possible through the central console. That was the motivation for us to move to one of these products. SentinelOne was our first choice, but we ran into a specific issue.
We had not specifically signed up for any risk management, but we were also looking to expand that to a completely managed SOC where we do the log correlation as well. When we initially started, we only started with the endpoint, identity, and cloud.
How has it helped my organization?
The main reason for getting this solution was that it was a new-gen endpoint solution for having an organization-wide view of security vulnerabilities or abnormal behavior. That was the main reason we got started with SentinelOne Singularity Complete. It gave us a lot of that information. It also helped us with compliance requirements. In the case of any specific instance or any abnormal behavior, its reports certainly helped us with the root cause analysis and collection of logs. It helped us in providing or collecting the evidence that we could use in our compliance reports to ensure proper reporting for relevant legal entities.
The ranger product helped us to do discovery of endpoints. We could identify our rogue devices.
SentinelOne Singularity Complete helped to reduce alerts. It groups the alerts. If you have similar alerts coming from the same server or a couple of servers at a similar time frame, it groups them and sends a single alert along with the device ID. This way, you have less number of alerts for the team to work on. If the agent itself is not in the running state or does not have the latest signatures available, it basically groups the alerts and tries to create a single alert. You have all the endpoints listed out, and you can take action against that particular issue rather than the same issue being reported from thousands of machines together. It is hard to provide the metrics, but generally, it helped quite a bit. I had around 8,000 endpoint licenses, and if 20% of the services started reporting the same issue, there would have been 1,500 to 1,600 alerts in a minute. It merges them into a single alert. We can also define a real-time action. A single alert helps our backend team to take action easily. The same is applicable to the SentinelOne support as well. If certain patches or certain actions are required to mitigate an issue, their team can do the mitigation in one shot and the fixes get pushed to all the servers that were reporting that particular issue. In one shot, you can automate and orchestrate your mitigation.
SentinelOne Singularity Complete helped reduce the mean time to detect and the mean time to resolution. There was at least a 10% reduction.
SentinelOne Singularity Complete did not help us save any direct costs, but there is an opportunity in terms of manhours saved in the backend because of having all these features integrated. There were indirect cost benefits. We saved a lot of hours because our engineers did not have to keep an eye on all the alerts. They could automate certain actions. That was an indirect cost benefit. I cannot list any direct cost benefits. These are costly products.
SentinelOne Singularity Complete absolutely helped reduce organizational risk. It is meant for that. We had different levels of reporting available. We could have an executive view. We could view the standards or framework that we were using. We could see the level of compliance to various standards in terms of percentage. We could also define the actions by accepting something as a risk or mitigating that by orchestrating.
What is most valuable?
There is centralized reporting and view. We can have role-based access management where technical people or monitoring people can have a central dashboard with a single view of all the endpoints. Whether our endpoints are running on Windows, Mac, Linux, or any flavor of operating systems, and even mobile devices, we can have a central dashboard through which we can do complete user management and policy management. We can have a complete security posture organization-wise, department-wise, or business-wise.
They have a good data lake kind of feature where you can ingest all the security logs. They can be from your endpoint, your identity management system, or your cloud. They can be from any of those services, so you get to do log analytics. That is one of the features that I liked about it. The same capability is also available with CrowdStrike which we are now exploring because of the issue with SentinelOne. However, at the time, with SentinelOne Singularity Complete, because of log analytics, we could do threat intel or sandboxing or have custom logic written for any specific kind of reaction. Those kinds of things were quite easy.
Log analytics and a couple of other things were also pretty good.
What needs improvement?
We ran into production issues related to CPU utilization on Linux endpoints. Our production environment's performance got degraded like anything. After a lot of debugging, we figured out that because it consumed a big percentage of the CPU and memory. Some of the applications were restarting automatically or randomly. We had an auto-healing infrastructure, so if the system memory was available, the application would restart on its own. When this issue got prolonged, we could see a lot of service failures because of being out of memory. This issue started hitting us wherever we had persistence connection requirements. Because existing connections were breaking completely, any transaction that somebody was doing online got terminated, and that was a big issue.
They should improve it for the open-source or Linux endpoints. They can provide customizations where we can limit the on-access CPU utilization or memory utilization. It should honor the specified limit and use only a limited percentage of CPU and memory rather than utilizing all the CPU or memory available on a system.
Other than that, I do not have any input. There is a lot of potential. There are a lot of possibilities for orchestration and sandboxing. Because we hit one particular issue, we were not able to continue using it, but I see a lot of opportunities there.
For how long have I used the solution?
With SentinelOne Singularity Complete, we did not work for a long time. We gave away this product within six months. There were some problems or issues reported, and that is why we discontinued using this product. We stopped using it nine to ten months ago. We have now migrated completely to CrowdStrike.
What do I think about the stability of the solution?
I discarded this product within six months. I would rate its stability a five out of ten.
What do I think about the scalability of the solution?
Its scalability is fine. I would rate it a nine out of ten for scalability.
We used it in a heterogeneous environment. We had about 8,000 endpoint licenses.
How are customer service and support?
I would rate their support a six out of ten because the issues that I had reported were not resolved.
As a strategic partner, SentinelOne is pretty good. They are very proactive.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
Prior to SentinelOne Singularity Complete, we had multiple pieces. We did not have one single product for everything. For endpoint security, we had McAfee as an antivirus and anti-malware. For identity, there was a different application altogether. For SIEM, there was a completely different solution, and for log correlation, we had a different log management server. Dashboarding solutions were completely different. EPO was the tool that we had to orchestrate some of the endpoint and antivirus-related policies.
We were having some challenges with SentinelOne Singularity Complete, so we migrated to CrowdStrike. We are now also exploring CrowdStrike's SIEM solution.
From a maturity standpoint, both SentinelOne Singularity Complete and CrowdStrike are mature products.
How was the initial setup?
We deployed it on-prem and on the cloud. Its deployment was straightforward. It was orchestrated via my backend tool.
It does not require much maintenance. The maintenance required is similar to an endpoint. One or two people are sufficient for 8,000 to 9,000 licenses because they need to just monitor the status. In case they find a rogue device, then only they have to take action. Otherwise, once they have a complete deployment done, they just need to automate reports and tasks. Those kinds of things certainly help.
What's my experience with pricing, setup cost, and licensing?
It is expensive. There is no doubt about it. If one of the functions does not work, it becomes very difficult for any CIO to justify the cost.
I would not be able to share the exact price, but we had almost 8,000 endpoint licenses, and it was a huge cost.
CrowdStrike is not cheaper than SentinelOne. Both products go neck to neck. Both are costly products.
What other advice do I have?
I would advise going for this solution only if you have a clear use case.
I have only one recommendation. If anybody wants to use such a solution to its potential, they need to be very clear about their use case. They need to know whether they want to go for the complete solution or they are just focusing on the endpoint solution. If you have a complete use case that requires EDR, identity, cloud, and log analytics, then SentinelOne or CrowdStrike makes sense. If you only have an endpoint use case, then these solutions do not make sense. It would not be a cost-effective deal.
After the complete endpoint deployment, you have complete asset visibility. We never used the life cycle management piece. We were just using the EDR feature.
SentinelOne Singularity Complete did not help free up the time of our staff for other projects and tasks. It has a lot of potential to do that, but we used it for a very short duration. Because of the issue we had, we did not continue using this solution. However, it has a lot of potential.
I would rate SentinelOne Singularity Complete a six out of ten. After they improve the product and their support, I may increase the rating. At this time, I cannot rate it more than six.
Disclosure: My company does not have a business relationship with this vendor other than being a customer.