Transcript for "3 phases, 10 steps: Implementing a SASE architecture":
Hey, everyone. Thanks for joining. We're gonna get started in just a few minutes. In the meantime, we have some information about this webinar on the slide here. Gonna let people trickle in a little bit more. Okay. Alright. Thanks everyone for joining today wherever you're calling in from. So today's conversation is gonna be about Cloudflare's recommended strategy on how to make progress in adopting SASE or secure access service edge as your networking and security architecture. So we're gonna explore how to get started, what use cases to consider, and how we've seen customers make progress with this type of modernization project, as well as some demos so you can see what things look like, within when implementing with Cloudflare. Our goal is to get you thinking about your approach so you can get into more depth with the with experts for follow-up conversations. I'm James Chang. I'm with our product marketing organization. I've been with Cloudflare for four plus years now, all aligned with our SaaS portfolio. I'm joined by my colleagues, Chris and Victor. I'll let Chris introduce himself first, then Victor. Hey, everybody. I'm Chris Lopez, technical marketing engineer for CloudflareOne. Hi. I'm Victor. I'm part of the TASA team. He's the the team in charge of on their attack. So Thank you for joining me. Awesome. Thanks. So before we kinda get really into the meat of our presentation, I'm gonna launch a poll question, and I'm curious to just get a better sense for where you are in your journey with SaaS. Are are you just learning about it? Are you planning some initial use cases? Are you actively piloting things? I'm gonna give the people the opportunity to respond to this poll. And in the meantime, just set some ground rules for our conversation. So please feel free to write questions in the chat. We're gonna reply to as many as we can in time. We might also take some, live. And while this poll is running, I'm just gonna start with some level settings. So for those, you know, I'm sure you've been exposed to SASE as a as a term. You've probably seen it, the conventions you've been to. But, really, SASE, your secure access service edge is about moving from that picture on the left side to the one on the right. On the left is that classic legacy castle and moat model with lots of on prem tooling. Instead, organizations obviously wanna shift a lot of their security and network activity to a cloud platform and control plane. And in the process, those organizations are able to adopt security best practice like enforcing least privilege, default deny, and all these things associated with zero trust, another concept you probably heard. So that journey, left to the right, can be pretty daunting. And so without prescriptive guidance, it can be hard to know what to prioritize, and analysis paralysis can really slow down your projects. So I am seeing I'm going to, close the poll, and I'm going to give, while those poll results are coming in, I'm gonna give Victor an opportunity to kind of set the stage for, the conversation we're going to have based on his experiences in our threat advisory solutions engineering team. Yeah. Thank you, James. One of our models to deliver some security posture on or supporting the security on most of our services are on my team, which is the on the attack team. We I am the, solutions engineer from this this group. We're confident, to have the ability to mitigate DDoS, both layer seven, layer three attacks on most of your architectures, on most of your services, on most of your deployments, even they are on public clouds, data centers, even if you're an ecommerce, retail, operational, technology, financial, and, of course, all all of our native cloud native solutions ensure, you know, compliant standards like PCI, NIST, HIPAA, several, and soxly. Most of the solution James and Chris are going to treat this this morning, are supported, by all of our security posture from our solutions. We are happy to help twenty four seven. Thank you, James. Awesome. Thanks, Victor. So I'm gonna close the poll and share it out. It seems like the vast majority of folks are still in that kind of initial learning phase, with kind of exploring SASE, researching solutions. Some are beginning to identify some initial use cases. Others are expanding into further ones. So I think this kind of session will be pretty relevant to the vast majority of you. So we're gonna share our recommended journey, specifically calling out three overall phases and 10 common steps we see customers take as they progress towards a security and networking modernization project. And with SASE, customers are really trying to achieve two goals. First is kind of reflected in that top row. They're trying to simplify their infrastructure and tooling by consolidating capabilities on one composable Internet native platform. And the second goal reflected in that bottom row is to gain visibility and controls and adopt those zero trust security policies between users, resources, devices, and locations. So let's start with phase one. We see a lot of customers start their SaaSy journey by modernizing remote access and adopting zero trust network access or zTNA solution as their primary approach. This has been a focal point since the pandemic as organizations have been trying to adapt to distributed and hybrid work. So a common starting point is clientless is, for organizations is doing things in a clientless way for a priority subset of users or applications. And we're gonna go deeper on these use cases in a moment. Over time, most organizations then deploy the the device client so they can scale their approach to more users and apps. And here, there there's an opportunity to layer on checks for multi factor authentication, device posture, and other contextual and risk signals, and getting that forward proxy visibility. And that third step for many, the long term goal is to replace their legacy appliances like their virtual private network or VPN, and so that they can move towards a more robust default deny identity based approach to access across all their environments, including non web apps and legacy private networks. So let's look at some very popular clientless access use cases where we've seen customers begin offloading traffic from their VPN before scaling up and taking on those additional steps two and three. The first is securing access for contractors and third parties who are often on unmanaged devices. This was an early step for Indeed. That's the world's number one job site, and they took this as their initial step before rolling out to 13,000 users and removing their VPN entirely app in just three months. Another common use case is securing privileged access for developers who need to reach sensitive infrastructure. This was the starting point for Canva, the Australian graphic design company, and that helped them make the decision to to avoid building out their own internal identity proxy and access management system and instead working with Cloudflare to scale their approach as they've grown. The third is simplifying IT integration during m and a activity. So think about making it easier to onboard newly acquired teams without doing a full network merge. This was the case for Delivery Hero. That's, you're, you know, one of these large delivery platforms based in in Europe. And so for them, clientless e t and a simplified how they were able to expand into new regions when they need to acquire brands, and post pandemic when they face a significant demand in their business. And last but not least, we see organizations experiment with new cloud based e t and a for apps when they're when they're moving their apps to the cloud as opposed to not trying to retrofit their on prem VPN. And so for the investment company EQT, clientless ZTNA help them move away from their on prem Microsoft Active Directory approach and, you know, avoid trying to retrofit their custom proxies in on prem VPN. Now all these companies have ZTNA deployments in the thousands now, and addressing these common priorities help them build momentum as they scale. So that should hopefully give you a flavor of the types of use cases organizations take on. I'm now gonna show, a a a demo, a very short demo on how we see a lot of organizations get started. So we see here a Wiki server. It's hosted in Docker, and it posts internal company documentation. It's listening over, port 3,000. And right now, it doesn't have any wait for users to access it via an IP or a host name. And so what we're gonna do is set up a tunnel connection from that server to Cloudflare's network. So when we go and create a tunnel, we are going to, name that tunnel, and this is gonna open up that connection, which will allow us to create policies and for users to to reach them. So we're gonna pick what environment, the server is in, we're gonna copy a piece of code that we're gonna run directly on the server via the command line. And so we can see that the, software has been installed successfully. This is just a lightweight daemon that opens up that connection, and we can see that that connector has been established when we go back to the Cloudflare dashboard. So now we're gonna name what, this tunnel will be a public host name. So this is gonna be wiki.skyskyflash.co. And when people call to that host name, it's gonna resolve through local host, four three thousand over HTTP. And so now when we when we put that into our browser, we are gonna be able to resolve to that company internal documentation page, which is great, but it doesn't have any security controls, in front of it. So that's our next step. And we wanna do this based on identity in line with your security practices. So we're gonna go into the dashboard again and see the different modes of authentication. These are the ones that we've already configured, including a number of corporate SSOs as well as one time PIN. We support a variety, all the corporate SSOs you can think of as well as anything that's OpenID compliant or SAML compliant, as well as social identities. And then even within those corporate SSOs, there's an opportunity to kind of synchronize via SCIM, so that you can base policies as you evolve your identity approach. So now that we have seen the different variety of identities that we can support, we're gonna create an application. So we're gonna call this our Wiki application, and it's we're gonna, associate it with our wiki.skyflash.co. Now we're gonna have the opportunity to, create policies around this configuration. So in this case, we are gonna keep things relatively simple, and we're just gonna allow authentication via our corporate SSO like Okta as well as one time PIN. And now based on those authentication, groups, we're going to create a policy allowing employees to access this Wiki server. So we're gonna configure some rules for this employee's policy. We want them to authenticate via, we're we're gonna provide access to employees who are part of that Okta group. And we're also, in this case, going to require a form of multifactor authentication. And this you can this can take a variety of different ways of of forms of NFA, including hard keys. But as you'll see in a minute, we're gonna keep things relatively straightforward, and you can see that the Wiki application has been added to the dashboard. So now when we go back to our, Wiki page and refresh, we are presented with our authentication screen. And so we're gonna authenticate as Alice within her username and password, and she'll be presented with a second form of authentication. In this case, it's via Google Authenticator, which, you know, will pull up on your mobile device and you enter your code, and you've and we'll be able to resolve back to the, Wiki page. So that should just give you a flavor of some of the common initial steps. This demo is gonna keep going. I'm gonna move on, but you can create different policies for contractors, for example, or any type of user group that you want to encounter. So that is hopefully just a quick, pre snapshot of what this first phase looks like around modernizing remote access. And so as you kinda think about what you wanna do and where you wanna go and where you are today, we want to just provide some framework to help you think about what is good, better, and best as far as maturity levels for user authentication and resource authentication. So we can all agree, I think, that with all that, the maturity levels at the far left are bad. Right? We don't want people to rely on usernames and passwords. We don't think that users should have access to everything. And so we think every organization should try and make progress and get to at least stage two such that access is based on identity. And at the very least, you get groups of authenticated users to apps based on certain IP ranges. And, of course, the aspiration is to keep moving right. So for users, that means enforcing MFA, device posture, doing things based on context, additional adding as many risk signals as you can to strengthen your posture. And for resources, we have seen this vary quite a bit. So some organizations are pretty comfortable with just doing authentication to private network IP ranges. And while this may not be exactly aligned with zero trust, you know, it's a step in the right direction. Ultimately, we understand that how far you wanna go on this continuum will vary based on a lot of the factors. And so your implementation team needs to agree on what is considered enough for a given phase of your project and on what timeline. And so we hope in working with, you know, clients to help think about where they wanna fall on the spectrum. Okay. So we've looked we've talked about what that first phase is around modernizing remote access. In phase two, organizations are trying oftentimes to make progress by modernizing and consolidating security controls onto one platform. So this means unifying visibility and different, enforcement across all sorts of different environments, web, email, SaaS, cloud, and private apps. And so here, people think about defending against cyber threats, protecting data, and replacing complex and costly point solutions. For example, in step four, organizations are looking to strengthen threat defense across their web and email inboxes. So this requires technologies such as secure web gateway or remote browser isolation or cloud email security. And in this step, organizations look to extend Cloudflare to protect outbound Internet access with DNS filters while then looking to enforce more granular controls with HTTP and browser isolation. Then they often think about deploying email security to stop attacks like phishing and business email compromise. Often, this is a step organizations take just adding Cloudflare on top of native capabilities that exist in Microsoft and Google. And then in step five, they think about prioritizing, mitigating the risk of data exposure, particularly across SaaS environments with technologies like CloudX's security broker and DLP or data loss prevention. And over time, of course, you'd wanna keep extending those types of security, controls across more and more channels. So I'm gonna be turning over to my colleague Chris in just a moment. Chris will join us back on stage. In the meantime, I'm going to launch another poll, you know, to just get a better gauge on what capabilities you are most curious, interested in deploying next. So this can be whether or not you don't have these controls today and or want to add them or whether you have them with another vendor, but your next project is to replace them. So I'm gonna run this poll in the background, and we're gonna keep going with the presentation, and we'll close out the results before our next demo in a minute. But we've seen a lot of organizations tackle these this phase of of of the journey in in a bunch of different ways. So there's THG. They're an ecommerce retailer in The UK. They had started down that path of replacing their VPN, but saw an opportunity to unify security across their web and applications across their hybrid workforce. So they got their migration started, and we actually had some automated tooling that helped them lift and shift some of their policies from Zscaler to Cloudflare to ease that transition off their existing vendor. Applied Systems, they're a US insurance technology. They follow the similar path. So for them, it was an they value the opportunity to consolidate web filtering and application access on one platform, and in the process, replace point solutions like their Cisco AnyConnect VPN as well as Zscaler for Internet filtering. Estonian Railways, one of their priorities was to get a better understanding of their SaaS security posture, And so they did this by deploying our CASB and doing easy and read only API integrations to figure out where there were potentials for data leaks. And then Werner Enterprises, this is a North American transportation company. They really focused on securing their email inboxes and were able to reduce their malicious emails by over 50% as well as cut, their the manual hours required to go through those emails and triage them by several hours a day just within the first few weeks of their deployment. So hopefully, this gives you a flavor of what, extending security in the second phase looks like with Cloudflare. And now I'm gonna turn it to Chris, who has a live demo for us around, deploying these different types of technologies. So, Chris, over to you. I'm gonna stop sharing. Alright. Thanks, James. I'm gonna share my screen real fast. Hey, everybody. So do we have any poll results in yet? Yes. So we yeah. Lot of folks are interested primarily in DNS filtering. Awesome. So that's actually where we are right now. So you can see here we're under gateway. This is all of our forward proxy policies. So we have a number of policy set up here like blocking high risk sites, blocking inappropriate sites. So if I were to just dive into this, this looks like your, you know, your regular, acceptable use policy, so content categories are in, you know, things like gambling, you know, questionable content, stuff like that, and ways it looks is if I go to something that would register as one of these categories, so poker.com, you can see that Cloudflare blocks it. And also, not only does it block it, but we get a very helpful coaching message saying that this was blocked because it violates our acceptable use policy, which if I go down here, to the bottom, I wanna point out that we can also have a specific display, from our device client saying that we have, you know, this this triggered for a very specific reason. So of the actions that are available within any DNS policy, you obviously allow and block. You have the ability to override it. So if you are matching to, if somebody is trying to go to a nonapproved or unsanctioned application and it matches to that at a DNS level, then you would be able to simply override that to a different host name. So if someone wants to go to Box, but they need to go to Dropbox, DNS policy would also be able to take care, of that, in addition to HDB policies. But you you also enforce safe search, and you'd also be able to, restrict YouTube access, with DNS policies as well. So with that, I'm gonna go ahead and switch over to HTTP policies. So here, again, this this happens at the HTTP level, and we're going to have a wider range of activities and wider range of of policies that we're going to be able to enforce. So I'll try I'll try and keep it to, like, the, you know, the greatest hits here. Obviously, we can still block high risk websites like we would do with DNS policies. But over here, you see we have an allow policy, but, really, this is just blocking access, to our company, Google tenant. Now the way this works is we just apply a header, down here. Custom header name is our big company skyflash.co. And whenever I try to go to, you know, any Gmail instance that is not specifically associated with skyflash.co, you can see that we get a, message back saying this service is not available. In addition, let's back up here. We also have the ability, to have granular controls, over things like activities on SaaS applications. What do I mean by that? I mean, over here, we have a policy that says stop unauthorized Salesforce record edits. And down here, you can see it's a bit more of a complex policy because we're specifying the HTTP method. We're specifying, you know, the URL that matches to an edit button. And we're also having some identity and potentially some device posture elements in here. And I wanna point out that here, you'd be able to say if you you could match all of the identity, you could be normally allowed to do this. You could match all the identity criteria. But if you don't match the device posture criteria as if as in, you know, if you don't have a device client active, if you don't have, you know, things like, you know, an active CrowdStrike client or, you know, other third party EDR, then you'd be denied from the policy itself. So over here, this is a policy that would stop unauthorized Salesforce record edits. So what that means is if I go here to, you know, an instance of Salesforce, I wanna point out that I have my device client active here, and I'm gonna try and just edit this. We'll say, not, you know, a 100,000 for this one, and I'll try and save it. You can see that we hit a snag. Not only did we hit a snag, but we also got a notification from our device client down here saying that we couldn't edit this. In addition, I'm gonna go here to a, document full of sensitive information, and I'm gonna try and download it. And you can see over here that it has failed. So it says check Internet connection, but that's really just the nature of how you create your network with Cloudflare. Essentially, you just on you on ramp all of your traffic to Cloudflare and as a result then, you know, that's that's where we apply all the security policies and as a result of that. The second that this list of customer employees, gets, you know, it traverses Cloudflare's network, matches to the policy, and we're able to zap it. So this also applies to things like, this also applies to things like AI applications as well. So we'll be able to inspect, you know, one it wouldn't just be files being uploaded and downloaded. But if I go over here data loss prevention, we'd also have, DLP profiles. I mean, you're, you know, just starting off with Social Security, you know, insurance and tax identifier numbers. Pretty pretty straightforward. But over here, we have prompts for, you know, AI prompts for security, AI prompts for financial information. So if you're going to, you know, a chat GPT or a cloud and you are putting sensitive information into the prompt, we're also going to be able to detect that. So over here, I do wanna point out that this we also have the ability in our policies here to, scan files. So we have a, we we do have a sandbox. So we quarantine all files from suspicious sites. Currently, the file, you know, we're testing this with, sandbox.cloudflowdemos.com. And you can see here, we can download a test file. And then, you know, obviously, here, our file is being scanned, and this is just a demonstration of what the process would look like, when we do this. And, also, finally and last but certainly not least is browser isolation. We do have, you know, we do have a very powerful browser isolation technology. It is, you know, it is ingrained, into our gateway service, but also you can do it you can use it as a as a, you know, as part of z t and a, as part of a policy, which I'll show you in a moment. But over here, you can see we are isolating social media, streaming websites, and new domains. What this means is, you know, I'm I'm looking at things like, you know, Facebook, Reddit, TikTok, and we're saying, if you you know, if if it matches or if your outbound traffic matches to any of this, we're gonna isolate that. And not only are we going to isolate it, we're also going to apply a number of data protection controls on top of it. So we're not allowing copy paste. We're not allowing printing or uploads and downloads. And to test this, I'm just gonna navigate to, Reddit real fast. You can see that in just a moment, yep, we have the browser isolation toolbar. So this is, basically, just a toolbar pointing out the, you know, the the time it takes for the isolated browser that spun up at Cloudflare's Edge to interact, with your native device. And, like I pointed out, we have the data protection controls in place, so the website printing, here is disabled. Copy paste would also be disabled, and if I attempted to upload or download files, that would also be disabled. So I was getting browser isolation through our client here. We also have the ability to do it without a client. And all of the policies that you would see in Git, you know, firewall policies, so DNS, HTTP, these would all work inside of our clientless stand alone browser. So what that looks like is if I go here to our instance, I get here to the browser and I'm brought simply to a web page that anything I search inside of here will be rendered in an isolated browser. So if I were to go to cnn.com, then, you know, this, much like you saw before, is, loaded in an isolated browser. So this is accomplished, using a SKIA draw channel or basically a SKIA draw message channels to be able to draw natively what's happening at the end destination onto your device. So your device never directly interacts with the, you know, with the service at the end of it. Alright. And that as a really brief overview, we could go deeper, but, just gonna in respect to time, hop over to CASB. So this is where you're going to be on and this is API based CASB. If you're looking for, you know, inline CASB, a lot of the inline CASB functions are going to be part of our zTNA solution to have their access, mainly because we're going to be able to apply the same access policies for a self host application, as we do for a SaaS application. Now this is more API CASB. So over here, we have a number of integrations. So, you know, Box. So you have, you know, Confluence, Jira, Dropbox, Slack. What's happening is we're enabling a number of service level accounts, you know, that Cloudflare has access to, and Cloudflare is going to perform a number of API scans and then indicate, what misconfigurations may or may not be present in your services. So that way, your your security team doesn't need to spend a lot of time looking for them. So over here, you can see, in box, we have, like, a file that's publicly accessible with view access and a DLP profile match. And if we dive into this, we, you know, we wanna manage this. We're gonna be able to see all of the files that are currently present in our instance of box, and also just, you know, the severity of this. So and it also would look like, you know, you have an active third party app with access to Jira, and it will tell you, you know, what apps, you know, necessarily have this access. And odds are there are apps that do need to have access, but it allows you to evaluate it by surfacing everything, and saving you the time of looking for it yourself. Also, as you can see here, CASB and gateway, are both going to work with our DLP policies or DLP profiles, and it'll be able to surface, any kind of, you know, any overexposed data or any sensitive information, that might be stored in the wrong location. So I guess just to bring it all back, you've seen, everything from DNS filtering, to be able to just, you know, do basic traffic blocking. You've seen HTTP and how we can, you know, not only, apply d you know, DLP policies to this, but also have fine grained controls, over your SaaS applications, and also tenant control, so that we can redirect people to approve corporate alternatives, and also browser isolation, and sandboxing, in addition to what you see here with our API based CASB. So with that, I'm going to turn it back over to James, and he'll be able to, continue the rest of the presentation. Yeah. Thanks, Chris. Actually, while we're on the topic of security, we've had some questions in the chat. Figured it'd be good to just, address live. So first question is, how would you allow access to just one specific YouTube channel, for example, with, you know, SWIFT policies? Only a specific channel? Yes. As long as you have the URL of it. Actually, I could just it'll probably be help, better with the visual aid. So if I were to say, hey. We want to make sure that we only have access to a specific channel, and I haven't tested this specifically. This is a brand new request. But as long as we have the URL of the channel, we'd be able to allow traffic into that. So we would say, like, if the, if URL, you know, we're gonna block YouTube and then we would layer a policy above this, that would say, but we're going to allow this channel of YouTube, and then that allow policy would strike first. But if they don't match to that specific channel, they get blocked. Yep. Yep. Just kind of a combination of the classic allow and blocking. So we also had some other questions about I think maybe in our earlier demo, there was a a selector for HTTP versus HTTPS, you know, traffic when we were setting up the tunnel and creating policies and so on. Just wanted to, you know, give us the give you opportunity to kinda clarify, you know, how do you set up tunnels via HTTPS. Got it. That's gonna be just as simple as, you know, setting the service type over here. And now, obviously, with HTTPS, there will be you know, you you will need to have your certificates to all those ducks in a row. Cloudflare does provide you, the ability to generate your own certificates or just upload your own if you'd like over here in resources. You'd be able to generate them and use them for, you know, whatever purpose you would you would really want to use them for. But if you do have HTTPS, then, again, you would be able to just set the service inside of the tunnel. Great. Thanks, Chris. Hopefully, that answered, some of the the questions, but happy to, you know, at least keep them coming in the chat. If if you after you stop sharing your screen, I'll reshare mine, and we'll keep going. Got it. So, okay. We have, so far, really been talking about, you know, modernizing remote access, extending security. This final phase is about shifting network connectivity and security to the cloud. So I think we can all fairly say that relying on routers, firewalls, MPLS lines, all those kind of traditional appliances really holds back business agility. So the goal here is to replace all that have overhead across your branches, your offices, your data centers, and other locations. So step seven is kind of reducing those kind of bulky networking stacks that say on prem in your offices. I think we're all familiar with the the experience of working from a WeWork or a hotel or a coffee shop. You kinda grab your spot, you connect to Wi Fi, and you get to work. And so companies are increasingly wanting to deliver these type of simple, really consistent experiences wherever their employees are and in a more lighter way, more cloud native manner. And so this is kind of the trend known as coffee shop networking. And so here, the SaaS platform, not the on prem firewall, is providing, the connectivity and the protection. With step eight, organizations are trying to manage different, you know, infrastructure across different clouds, across AWS, GCP, Azure. They wanna do with greater consistency and flexibility. So for example, Cloudflare helps organizations automatically discover and manage data functionality of their cloud resources, like their virtual private clouds or v p VPCs, subnets, virtual machines, route tables, routes, all those types of things without requiring any additional two cute, VMs. Eventually, in step nine, the goal is to eliminate all those types of private network circuits, especially those, you know, sticky MPLS connections, and instead route all site to site traffic through the SASE platform. And as you you can always get more granular with your zero trust approach, so your networks, your environments, all those things are gonna continue to expand. And so that final step 10 is just to continue refining those policies to be even more adaptive to context and evolving risk factors. So, you know, Chris, I think this is a good opportunity to kind of highlight the, variety of different on ramps that exist. Absolutely. So as you can see from James' screen here, we do have a number of different methods to on ramp your traffic to Cloudflare. And, again, that is how it works. You have to get your traffic to us. We're gonna be perform all the necessary network and security services on top of that. And here's how you do this. So what you saw in the demo was me turning on a device agent, so that's automatically going to, you know, create a new network stack here. It's gonna create an encrypted tunnel to Cloudflare, and that's where, you know, that's how we on ramp and that's how the policies get applied. But there's obviously gonna be a lot of instances where we can't have devices on your device clients, on everybody's device, and in that in that sense, you'd be able to see, like, a WAN location down here at headquarters. Right? So we have, like, a WAN connector, for a retail location. So that's just, you know, IPSec tunnels configured via, you know, two firewalls, back to, Cloudflare. And then over here in headquarters, we do have a work connector. So that is a software based, app connector that you would then also be able to use, to the same effect as you would a, you know, the WAN connector. Up here in cloud, I do wanna point out that we have other instances of the WARP connector. Again, software based, very easy to use. And then over in, Azure, you can see that we have Cloudflare d, which was the original, it you know, similarly to WARP connector, it is a software based, application connector, but it, is a bit more lightweight, and as a result, it's gonna be better for just, like, web based applications, things that don't involve bidirectional communication, like, you know, unified communications, streaming, gaming, that kind of thing. And then over here, you do see that we have a transit gateway, for AWS. So we obviously recognize that you're going to have other services, or, you know, other similar services with, like, your hyperscalers and things like that. Whatever you have, inside a a hyperscaler that you're currently using, Cloudflare is going to be able to work alongside that, very agnostically and very effectively. So that way you don't have to have, you know, large dramatic changes, to your infrastructure, in order to, you know, in order to onboard us. Excellent. Thanks, Chris. We actually had another question in the chat. Are we correct to assume that there's an API to do things programmatically? Yes. We have one API across all Cloudflare services. Chris interacts with it, you know, often to do a lot of orchestration. And one of the things that, you know, customers really value is the kind of simplicity of just having one API to interact with all of Cloudflare such that they can do things programmatically and automate. But, Chris, anything you wanna add to that? Yeah. Honestly, Cloudflare isn't they they build the API first and they build the GUI on top of it. Also, we do work alongside, like, the open source orchestration platforms like Terraform. And you can use that very easily to spin up an environment to the greenfield or, you know, make additions brownfield. Awesome. Thanks. So just kind of moving to wrap up. Right? We talked about having three phases overall in kind of a SaaS journey. But, ultimately, customers can and have prioritized starting with any of these three phases and any of the specific steps or use cases within that SaaS journey is gonna be up to you and your priorities and where you are in your overall modernization, the types of IT that you currently have and and your aspirations. So if you wanna go deeper on implementation guidance, we really recommend the SASE reference architecture. It includes a lot of technical and steps and best practices, some of which, Chris alluded to here, but you can go much deeper, in the this QR code. Or when you talk to our team, there are a bunch of different ways to get started, whether that's through architecture workshops where we can see where you are today, where you wanna get to, different forms of proof of concepts and trials, as well as business value assessments to help you make case internally for your project. So while we're doing this, we have a a a couple of additional questions in in the chat, which I'll I'll you know, we'll we'll take live. So, I have a question about the warp connector, Chris. Is it meant to be kind of an onboarding or a placement for site firewalls? Just maybe this is an opportunity to talk about, you know, our decision to, you know, offer Warp Connector as a method of onboarding traffic. Yeah. So, Warp Connector, it came about because, Cloudflare, one one of its limitations is that it won't be able to deal with bidirectional traffic. So the Warp Connector was brought out, to be able to handle those use cases. As a result of it being so new, it just has, you know, less support for different OSes, and it's more of a you know, if you if you have this specific use case, you should use WARP connector. If you have unified communications, if you have an environment like CICD that requires bidirectional communication, then we have you know, we're able to cover you. But, typically, customers would go for Cloud thirty first. And if you do need a firewall, we do, offer, like, you know, firewall as a service. But, again, that would all be software based. It wouldn't necessarily be a a hardware replacement. Great. So I just answered another question in the chat. We are gonna be around a little bit longer for questions, but, otherwise, thank you so much for joining. This is a great opportunity to download some additional resources. If you go to the docs section on the right hand side of your screen, we have an ebook that kind of goes through these 10 steps and these three phases that we, highlighted. There's an infographic so you can get get a sense for how other customers have adopted Cloudflare for these types of use cases. And if you wanna go a little bit deeper, there's, of course, that reference architecture I, shared in the prior slide, as well as a just a brief on our overall SaaSy platform. But thanks so much for your time, everyone. Do you we have a link for upcoming sessions. I think that's maybe better best to ask by Salaam or maybe you can answer answer in chat. But but, yes, we will get that to you. If there are no other questions, thank you so much for joining today, and hope to see you at 10 future security builder workshops. Thanks a lot. Alright. I'm leave. Bye.