Navigating the Complexities of Hybrid and Multi-cloud Environments

Ever since AWS launched Elastic Compute back in two thousand six, the adoption of cloud technologies has transformed drastically. Now initially, the push was for all in cloud strategies. But over time, it became clear that a combination of on-premises and cloud resources or hybrid cloud was more effective for many organizations for a variety of reasons. And then soon after, we saw organizations leveraging multiple clouds and SaaS providers right alongside their hybrid cloud architecture. And this is the problem that today's engineers face. Complex architectures that mix public cloud, SaaS providers, private data centers, containers, network overlays, and everything in between. This affects how we approach performance monitoring, security, compliance, and how we track costs. Kentik Cloud takes a unified approach to help engineers see, understand, and manage the entire cloud network landscape. Kentik ingests data from across hybrid and multicloud environments as well as on-premises data centers, containers, and even service provider networks so that you could see how application traffic is moving between regions, among public clouds, and over the public Internet. So let's take a look at an overview of how the Kentik platform works to see the kind of hybrid and multicloud visibility we provide. Okay. Let's start with the Kentik map where we see an overview of our entire cloud environment, including multiple public clouds if you're using them. And we can drill down in pretty much any direction that we want from here. So for example, if you wanna look at AWS specifically, you could do that from here if that's all you need, but that's really just the beginning. From here, you can really dig into your cloud topology. In this case, we're gonna look at AWS and really understand how traffic is moving among regions, through transit gateways between VPCs, and between your on-premises resources and your public cloud resources. And notice here at the top, we have our on-prem routers. So keep in mind that we're ingesting information from those on-premises network devices into the same platform as cloud telemetry, which means you get an end to end picture and you can pivot to look at maybe a relevant WAN interface on a branch office router or data center firewall, whatever is important to you. Or just as easily look at a direct connect or a transit gateway which you can see here on your screen. And this is a very important visualization to have if you're running a hybrid network. So we could see here, from this router, we're going through this direct connection and we're heading through this gateway. And if I scroll down for you, you could see exactly which transit gateway we're using for this AWS region. So let's open up this region and now you could see the traffic moving among and between our VPCs. And most of what you see here on the screen is clickable so you can keep drilling down even deeper and you wanna see more specific information like, specific IP addresses, allow and deny statements, and so on. So now I'm gonna shift to the data explorer to get even more specific with what and exactly how we can filter all of this information that we're ingesting into the system. So I'll open up data explorer and I'm gonna set up a query for looking at VPC to VPC traffic in AWS, just as one example. So all I'm gonna do is select AWS from our data sources that we also support other public clouds. From our dimensions, which is our filtering options, I'll select VPC. And then why don't I also grab IP addresses? We'll run the query and the result is that we're looking at VPC to VPC traffic which we can show in other various visualizations as well. So I'll change this to a Sankey just so you can see your options. Now based on all these filter options, which again we call dimensions in the platform, you can get very granular, very specific in network and cloud visibility stacking the information that you wanna see from the specific time range that you wanna see as well. So whether you wanna filter by things like application, site tags, DNS geolocation, you can combine filters together, whatever works for you for your specific use case. So now let's say we wanna filter for cloud to on-prem traffic. I wanna see what a specific VPC is talking to on-premises. So I'll select the cloud traffic that we wanna see, which you can see here are all the major clouds, but I need to get more specific. So I can choose the specific VPC that I want. Now for this demo, I'm gonna select VPC seventy which I have set up in AWS as part of this demo environment. And I wanna see traffic going back and forth between that VPC and my on-prem network. And I can filter for specific devices, site locations, IP addresses. But in this case, what I'll do is leave it blank just to see what this VPC is talking to back on-prem. So I'll select source, destination IP, and why don't I grab the application as well. We're gonna run this query and now we can see my traffic from VPC seventy, which is the ten dot seventy network. And we can see the traffic now between it and this subnet in particular, it looks like, the ten dot one thirty subnet, which I know is an on-premises network. And we can change this to a Sankey if that helps or whatever we want. And you can continue to drill down on these individual IPs, create new filters, create new queries based on just one of these line outputs. And you can also generate a report if you need to right from here. Now to get an idea of network performance to one of the public clouds, we can look at the state of the Internet, which is built on Kentik's global collection of synthetic test agents. But you can certainly run your own test agents from whatever vantage points make sense for your organization. And as you see here on your screen, it seems that we do have a performance problem between two AWS regions. And if I click on details, we can look at that information, see what's really happening both from a metrics perspective and what's happening in the path. So, basically, we're gathering as much information as we can from public cloud, test agents, on-premises resources, and collecting it in one place for an engineer to both visualize and then explore. Stay ahead of visibility with Kentik because in the world of cloud computing, visibility isn't about seeing just one piece of your environment. It's about seeing how it all operates together so you can troubleshoot, plan, make decisions with precision. For more information, visit kentik.com/cloud or reach out to your local Kentik representative. Thanks.