How has it helped my organization?
Awake has made us more productive. We're spending less time looking at false positives, so we can focus on what's truly important. It hasn't affected the morale of our analysts because we use a third-party SOC.
When I look at the central dashboards, I can see what adversarial models were matched within the day, and when I click on that day, I can see what models and device names got triggered within my homepage. If I want to dive further into that model, I can click on that, and it tells me what the threats were as well as a lot more information on the endpoint or the asset. Then, if I want to see even more information, such as the actual activities, it's three clicks, and I'm on the activities themselves. I can pull a PCAP and investigate it. Regarding responsiveness and how quickly I get the answer, it's much faster than what I used to have.
It's hard to quantify, but it would have taken me 10 minutes to figure it out in my previous solution because I'm on the platform every day. Awake is easier and more intuitive. You see the day, the triggered models, and the asset. Then you click on the asset and activities. They're right there. I get the source, destination, and details, then download my PCAP, and I'm done.
Awake also tracks unmanaged devices. We have a guest WiFi, so if someone logs in to that, it's an unmanaged device. If they log in and try to do something bad, Awake will flag it and tell me. It's important even though we don't have as many people coming in and using the guest WiFi due to COVID, but we need to know if a guest user is doing something malicious.
What is most valuable?
It's much easier to create your own queries and hunt for threats. Darktrace's language is more challenging, and it's almost like you have to learn Darktrace's methodology to decipher it. When I create a workbench query in Awake to do threat hunting, it's much easier to query. You get a dictionary popup immediately when you try to type a new query. It says, "You want to search for a device?" Then you type in "D-E," and it gives you a list of commands, like device, data set behavior, etc. That gives you the ability to build your own query. Gathering PCAPs is also quite practical and more straightforward— tweaking the adversarial models, too. With Darktrace, it was tough to do. If you go to another serial model and want to clone it, then edit it and disable the old one, you can do it easily.
We have Palo Altos to decrypt traffic. I have all traffic going in and out via Awake, which can decrypt the traffic. However, Awake doesn't need to decrypt because it can analyze encrypted traffic to get a sense of what it might be. What I find helpful is that Awake can tell me when encrypted files might contain passwords. There is an adversarial model for that, which is great when someone tells me that there are two files with passwords, but the Awake and DR team already has an open ticket for this. They look for files that have "passwords" in the filename.
That allows me to reach out to the user and tell him that I noticed a file containing passwords, and it's not password-protected. When they password-protect the file, the Awake team still highlights that as a risk but then write to them and say a password now protects the password file, and even though it is a password file, it is encrypted. So if you try to open it, you have to decrypt it with a password. Then we tweak the model to prevent that model from being triggered for that specific filename.
What needs improvement?
We take in IOCs from my SOC and from AlienVault, and then we focus on traffic that hits IOCs and alerts us to it. The one thing that the Awake platform lacks is the ability to automate the ingestion of IOCs rather than having to import CSV files or JSON files manually. Awake didn't support the manual importation of CSV and JSON in version 3.0, but they added it in version 4.0. It's helpful, but it still has to be a specific CSV format. Automated IOCs are on the roadmap. Hopefully, they will be able to automate the ingestion of IOCs by Q1 next year. I'm currently leveraging Mind Meld, an open-source tool by Palo Alto, to ingest IOCs from external parties. I aggregate those lists and spit them out as a massive list of domains, hashes, file names, IPS. Then we aggregate those into their own specific categories, like a URL category. Awake ingests that just like the Palo Alto firewall does, and then it alerts me if traffic attempts to go into it.
Some of that is already on the Palo Alto firewall, which blocks it, but that doesn't mean that there is no attempted communication. I want to know if there's a communication attempt because there might be an indicator on that specific device trying to reach an IOC. Yes, my Palo Alto blocked it, but there's still something odd sitting there, and what if it can reach a different IOC that I don't have information about? I want to focus on it. I could do that by leveraging Awake if it could ingest the IOCs automatically. That's something I leverage Awake for today. I still have to manually import it, which is cumbersome because I have to manipulate the files that I get from the different IOC providers into a specific format that it understands. Once they add the ability to automate that, it'll be more useful.
For how long have I used the solution?
I have been using Awake since 2020. They hadn't been acquired yet by Arista when I joined.
What do I think about the stability of the solution?
Awake is pretty stable. It has come a long way. There were quite a lot of bugs initially when I had them in version 3.0. I'm on 4.11 now, so it's a lot cleaner, more intuitive, and much less buggy. I found bugs as each new release came out. I brought them to the attention of support, and they would fix them, then I'd find a different one. I can't comment now since Arista acquired them, but before Arista, the development to get something fixed was much faster.
What do I think about the scalability of the solution?
I have a larger appliance than I technically would need, but I prefer that. If my organization goes up 100 percent, the appliance will still be suitable. So the scalability is there. If you switch from a 50-person shop to a 1000-person shop, it's easy to upgrade the appliance. They get a new one, install it, migrate the data, and you're done. I don't have any reservations about that.
How are customer service and support?
I don't think anyone is a 10 out of ten. There's always room for improvement. I'll give the Arista support group an eight out of 10, and nine and a half to the MNDR team. Awake's managed network detection and response service is fantastic. Awake MNDR has been there night and day for us. In fact, they've helped me a couple of times where my SOC has fallen short. They got me the answers I wanted, which is precisely why I wanted to sign up for MNDR.
Awake MNDR has made our security posture more comfortable. We get some peace of mind knowing they're there if something should happen. I can reach out, and also, they open their own tickets for things they see that the Awake platform doesn't necessarily catch automatically. You want that human element behind it, not just the EML component of it, where you build these models as an ML. You tell the machine what to look for, and if the machine sees it, then it tells us something about it. It's not machine learning — more like machine finding. These guys are looking for the nuances that the machine can't find.
If they see new IOCs, attack vectors, methods of attack, hashes, or techniques, they're going to log in to random customers and do some threat hunting. We get a lot of value from having the ability to say, "Guys, I heard about X, Y, Z. Can you check if there's any indication of that in my environment?" They can then log in, do their own threat hunting, and tell me, "No, categorically, there isn't." That's a lot more helpful than just having a SOC.
If my SOC is spending a couple of hours doing it, they're not going to be Awake experts, of course, because they're a SOC, and they probably have to leverage so many security tools it's impossible. They all have customers with Vectra, Darktrace, etc., and you can't learn them all. So having the Awake team allows me just to ask the Awake MNDR team, "I got this ticket. Can you guys log in and investigate it?" Or, "I have this question. This user did XYZ. Can you guys investigate this and paint a picture based on what you see in Awake." Of course, they don't have access to SentinelOne or a lot of my other tools like the SOC does, but they can give me a sense of exactly what happened just by leveraging Awake.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
Previously, we were Darktrace customers, and we had the Darktrace platform set up in two locations: here and our data center. We leveraged them because we wanted to have an NDR solution. Darktrace is great eye candy, but we got a lot of false positives in the environment. When we spoke with Darktrace, they assured us that it was AI with machine learning capabilities so that it would adapt to our environment the longer it was deployed.
I'm not sure if they've gotten better since then because I left them two years ago, but our SOC was spending too much time looking at false positives. When we approached Darktrace and told them that the solution was flagging functions that were normal in our environment, the support was not up to scratch. If you constantly have to change the model and tell it to ignore issues in your environment, then that's not machine learning because it's not learning the environment.
Awake had what I was looking for with Darktrace but didn't get, which was to get a response. So you detect it and respond to it by integrating it with the EDR tool, specifically at the endpoint. I wanted a response, but that automation wasn't there. Darktrace has it now. However, Awake had the EDR integration to Crowdstrike and SentinelOne out-of-the-box, which was great because then I wanted to do it, but it's not fully automated yet. I can isolate the endpoint from the Awake platform but there's still no playbook yet where it says, "Okay, if you find a ransomware attack going on, isolate that endpoint and respond automatically." That's on Awake's roadmap.
Another reason I moved to Awake was that they're not truly an ML or AI, and they don't sell themselves as that. They look at it differently from a security perspective, and I like that. The integration with EDR is better than what I had. They were looking to integrate with Palo Alto and Cisco firewalls to automate the response to IOC. If an IOC is identified in my environment, it will tell my firewall to start dropping the traffic to the IOC. They don't have this functionality yet, but I know it's in the roadmap because I just had a call with them about a month ago. I have a Palo Alto firewall, and the integration with Palo Alto will come along in Q1 next year.
I think Darktrace has this, or it's in the process of adding it, but Awake already had it on the roadmap two years ago. That was something they were building towards. Since then, I have expanded my relationship with Awake Arista by signing up for their MNDR service, which has been super helpful because we still get false positives when I tweak the adversarial models to match my environment. I don't think there's a solution that will genuinely learn your environment and know what's normal versus what's not. I've found that dealing with support is better than dealing with Darktrace. Granted, I have the MNDR team also now, but this was the case even before that. With the MNDR team, I send them an email telling them the alerts we've gotten and the workbench queries we used. Then I ask them to tweak the model, so we don't get false positives. After an hour or two, it's done. Compared to Darktrace, the level of responsiveness from Awake has been night and day.
I get low-risk false positives, and I treat them all the same, but I have a managed external SOC, and they will not. I do because I want to see less noise, and I want my SOC to focus on what's important. As such, I want to tweak the adversarial models to focus more on aspects that warrant research and response rather than just an alert that comes in. We can decide to look at something later when we have time because we can see it's a low-level risk. Awake categorizes these, so you know it's low when you see an alert with a risk score of 20. Still, I want to clean it up, so that I don't see them. When I look at my platform dashboard, I want to know that I have had X unique adversarial models for the past week and Y high-risk devices. Then I can zero in on those high-risk devices to see what they are and what they're doing.
I was a Dell Secureworks customer for a while. They were great tools, but they weren't NextGen. I thought Darktrace was NextGen. I had probably done a demo with them two years before becoming a client. I had Secureworks as a SOC, but then I wanted something more. When it was time to change my SOC from Secureworks, I figured I could use Darktrace and get an external SOC to ingest all of my security logs for the same cost I'm paying Dell Secureworks.
I thought that my SOC was spending too much time investigating all the false positives we were getting out of Darktrace, and it wasn't their job to tweak Darktrace. It was certainly more challenging for me to do it and more brutal to me to work with support to do it. And so, after attempting that for six months, I came across Awake. I can't remember exactly where. It must have been a marketing email I got, and I decided to look into it.
I think they had just come out of stealth mode when I started talking to them, and I decided to put them in at the same time I had Darktrace and do a bake-off. I realized that I was getting fewer false positives but, unfortunately, the platform does not have 3D manipulation, which I call the "eye candy" of Darktrace. It's an excellent visualization tool. It looks fantastic, but it's not easy to dive in and look at the logs.
I like how Darktrace can replay the traffic and show the messages coming in. I thought that was a pretty cool feature that I wish I could do with the Awake. But again, it's eye candy. The information is there, but you can't play it to the second as the traffic comes in. When I tried out Awake, I was taken aback because they had the IOC ingestion and were planning on automating that. They were also planning on integrating Awake with Palo Alto firewalls. Awake also had the EDR implementation as I was looking at migrating from Cylance to Crowdstrike. They already had Cylance integration also. I thought it was a no-brainer as long as I could get it for the same cost as Darktrace. I knew I would get a little more value out of it. I would lose the eye candy and the playback, but my SOC will spend less time looking at false positives.
I don't pay more or less if my SOC gets a thousand tickets or 10, but I also don't believe in my mailbox getting spammed with issues that worry me. Of course, I still get false positives from Awake. At most, it's maybe one a day, which is not terrible. We used to get five, but then I started tweaking it, and now we're getting roughly one every two days. We used to get five a day because no platform is built for your environment. They're built for all environments. They have to look for issues they think are malicious. You get that with SentinelOne too. I get false positives with SentinelOne and Excel files that look like they're meeting a MITRE ATT&CK framework, but they're not.
I think people should be ignored if they tell you there is a tool out there that's truly going to learn your environment. Darktrace claims that the tool will self-adjust the longer that it's in your environment. It won't. I've seen it, and unless that's been massively improved, I don't believe it.
What's my experience with pricing, setup cost, and licensing?
I got a deal when I bought Awake. It's if you go to buy a car and end up ripping off the dealer. I don't think many customers got the same deal. Darktrace is way too expensive, and so Awake is more price competitive. I think they'll be able to take a lot of clients from Darktrace because it costs a lot of money. All of these vendors push for four-year agreements and offer discounts for that. Darktrace told me that they only do four-year contracts, but I said I wouldn't be a customer if those were the terms. Instead, I got a four-year agreement with a 12-month opt-out. It's still a four-year agreement, but I could opt out after 12 months with a 90-day notice. So to me, it's a one-year agreement. I was able to get that with Darktrace because they wanted me as a customer.
Because I represent a hedge fund, I have some leverage. I told them that they had to meet my conditions if they wanted me as a client. It was the same way with Awake. They wanted an initial four-year agreement. Initially, we signed on for a one-year contract, but they wanted the four-year deal when it came time for the renewal. I told them that I was not doing that. I said that they either had to do it on my terms, or I'd go somewhere else. I don't want to, but I'll go.
We were able to keep the same conditions that I had, and working with them was pretty easy. I didn't have to jump through many hoops to get what I wanted. I was one of their first clients in the alternative investment space, and I've been a big supporter of what they were doing even before Arista bought them. I was worried when Arista bought them. When a conglomerate company bought this unicorn, I was afraid they would turn it into garbage.
Thankfully, I haven't seen that. The platform is improving, and the development continues. They're doing many exciting improvements that were on the roadmap when I first signed on. I can't disclose some of these improvements, but seeing what's coming down the pipeline is exciting. And like I said, I was fearful of Arista. Now I'm thankful that Arista pumped money into it and kept the team together, did not break them, that they're integrating them to their support model, and the teams will become bigger. And obviously, the interaction with the Arista products will become even larger because they're an Arista company, and they want to apply that to their Arista products.
My other big concern was that once Awake was acquired by Arista, they would have no interest in integrating with Palo Alto and Cisco because they are competitors. The sales rep told me, "No, that's incorrect. We still want to integrate with them. However, we understand customers are always going to have a choice, and not everyone chooses Arista for networking." I don't think Arista even does firewalls, so they put me at ease.
What other advice do I have?
I'd rate Awake Security Platform nine out of 10. I have recommended them to many of my peers and have done references since. I believe in Awake and what they're building. I know how much more they can do with this. Unlike Darktrace, Awake has been built from the ground up. Darktrace took a lot of open-source tools and integrated them. It may have been a sales pitch, but my understanding is that one aspect that sets Awake apart is that this platform is built from the ground up. They didn't take an open-source tool and bandaid it to another one to create a product.
That's one of the most exciting aspects of Awake. They can do what they want with this. They can build all these features on top of it. I bought into Awake because I wanted to get these features on a single platform. I want to create playbooks. I want something that can automate playbooks and leverage API calls to connect to your Palo Alto firewalls and SentinelOne. It's all about APIs nowadays. I want to have the ability through a single pane of glass that has your top 10 adversarial models that are critical. If you hit this criticality and you are up to this percentage, the following action that the Awake platform takes is X.
I believe that's where this platform can go, and I don't think any platform out there is at that level yet, even though Darktrace now has integration with EDR. They can automate many aspects, and they have added Palo Alto to it since then. Also, they have an email phishing component. I think Awake has the potential to do much more and based on the roadmap that I've seen, I believe they are well-positioned to do even better.
Which deployment model are you using for this solution?
On-premises
*Disclosure: PeerSpot 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.